WO2008018550A1 - Système de recommandation de programme, terminal de consultation de programme, programme de consultation de programme, procédé de consultation de programme, serveur de recommandation de programme, programme de recommandation de programme, et procédé de recommandation de programme - Google Patents
Système de recommandation de programme, terminal de consultation de programme, programme de consultation de programme, procédé de consultation de programme, serveur de recommandation de programme, programme de recommandation de programme, et procédé de recommandation de programme Download PDFInfo
- Publication number
- WO2008018550A1 WO2008018550A1 PCT/JP2007/065632 JP2007065632W WO2008018550A1 WO 2008018550 A1 WO2008018550 A1 WO 2008018550A1 JP 2007065632 W JP2007065632 W JP 2007065632W WO 2008018550 A1 WO2008018550 A1 WO 2008018550A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- program
- information
- user
- unit
- recommendation
- Prior art date
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N7/00—Television systems
- H04N7/16—Analogue secrecy systems; Analogue subscription systems
- H04N7/173—Analogue secrecy systems; Analogue subscription systems with two-way working, e.g. subscriber sending a programme selection signal
- H04N7/17309—Transmission or handling of upstream communications
- H04N7/17336—Handling of requests in head-ends
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/20—Servers specifically adapted for the distribution of content, e.g. VOD servers; Operations thereof
- H04N21/25—Management operations performed by the server for facilitating the content distribution or administrating data related to end-users or client devices, e.g. end-user or client device authentication, learning user preferences for recommending movies
- H04N21/251—Learning process for intelligent management, e.g. learning user preferences for recommending movies
- H04N21/252—Processing of multiple end-users' preferences to derive collaborative data
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/20—Servers specifically adapted for the distribution of content, e.g. VOD servers; Operations thereof
- H04N21/25—Management operations performed by the server for facilitating the content distribution or administrating data related to end-users or client devices, e.g. end-user or client device authentication, learning user preferences for recommending movies
- H04N21/258—Client or end-user data management, e.g. managing client capabilities, user preferences or demographics, processing of multiple end-users preferences to derive collaborative data
- H04N21/25866—Management of end-user data
- H04N21/25891—Management of end-user data being end-user preferences
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
- H04N21/45—Management operations performed by the client for facilitating the reception of or the interaction with the content or administrating data related to the end-user or to the client device itself, e.g. learning user preferences for recommending movies, resolving scheduling conflicts
- H04N21/4508—Management of client data or end-user data
- H04N21/4532—Management of client data or end-user data involving end-user characteristics, e.g. viewer profile, preferences
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
- H04N21/45—Management operations performed by the client for facilitating the reception of or the interaction with the content or administrating data related to the end-user or to the client device itself, e.g. learning user preferences for recommending movies, resolving scheduling conflicts
- H04N21/466—Learning process for intelligent management, e.g. learning user preferences for recommending movies
- H04N21/4661—Deriving a combined profile for a plurality of end-users of the same client, e.g. for family members within a home
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
- H04N21/45—Management operations performed by the client for facilitating the reception of or the interaction with the content or administrating data related to the end-user or to the client device itself, e.g. learning user preferences for recommending movies, resolving scheduling conflicts
- H04N21/466—Learning process for intelligent management, e.g. learning user preferences for recommending movies
- H04N21/4667—Processing of monitored end-user data, e.g. trend analysis based on the log file of viewer selections
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
- H04N21/47—End-user applications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
- H04N21/47—End-user applications
- H04N21/475—End-user interface for inputting end-user data, e.g. personal identification number [PIN], preference data
- H04N21/4755—End-user interface for inputting end-user data, e.g. personal identification number [PIN], preference data for defining user preferences, e.g. favourite actors or genre
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/60—Network structure or processes for video distribution between server and client or between remote clients; Control signalling between clients, server and network components; Transmission of management data between server and client, e.g. sending from server to client commands for recording incoming content stream; Communication details between server and client
- H04N21/65—Transmission of management data between client and server
- H04N21/658—Transmission by the client directed to the server
- H04N21/6582—Data stored in the client, e.g. viewing habits, hardware capabilities, credit card number
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/80—Generation or processing of content or additional data by content creator independently of the distribution process; Content per se
- H04N21/83—Generation or processing of protective or descriptive data associated with content; Content structuring
- H04N21/84—Generation or processing of descriptive data, e.g. content descriptors
Definitions
- Program recommendation system program viewing terminal, program viewing program, program viewing method, program recommendation server, program recommendation program, and program recommendation method
- the present invention accumulates user operations such as viewing a program, recording a program, playing a recorded program, erasing a recorded program, and recording a program, and provides the user with a reference based on collaborative filtering processing.
- the present invention relates to a program recommendation system for recommending a program, a program recommendation server, a program recommendation program, and a program recommendation method.
- the present invention also relates to a program viewing terminal, a program viewing program, and a program viewing method for recording or reproducing a received program.
- channels that can be selected by the user as television viewing forms are diversified, such as cable television broadcasting, satellite television broadcasting, digital television broadcasting, and IP (Internet Protocol) broadcasting that are not limited to conventional television broadcasting alone.
- IP Internet Protocol
- Deviation methods a vast amount of user viewing history information is collected, and users with the same preference are grouped by V, so-called collaborative filtering, and users within the group are identified. It is possible to extract recommended programs. Therefore, it is possible to provide a system that is advantageous to the user without the complicated operation of selecting a favorite program for the user.
- the second point is that only specific news and specific scenes in a program cannot be recommended, such as a news program. As a result, even if a program is recommended, it cannot be determined which scene in the program is recommended by the user. Eventually, the user needs to view and judge all the programs by himself / herself, which leaves troublesome processing for the user.
- Patent Document 1 Japanese Patent Laid-Open No. 2001-298677
- Patent Document 2 Japanese Patent Laid-Open No. 2005-33600
- the present invention has been made to solve the above problems, and generally has a high degree of preference.
- a program recommendation system is connected to a program viewing terminal for viewing a received program and the program viewing terminal via a network so that they can communicate with each other, and a user operation on the program
- a program recommendation system comprising: a program recommendation server that obtains operation information representing the information from the program viewing terminal and creates recommendation information related to a program that matches the user's preference according to the operation information.
- the viewing terminal includes an operation receiving unit that receives input of the operation information, and an operation information transmitting unit that transmits the operation information received by the operation receiving unit to the program recommendation server, wherein the program recommendation server includes: The operation information receiving unit that receives the operation information transmitted by the operation information transmitting unit, and the operation information received by the operation information receiving unit Based on the operation information storage unit stored for each user, the operation information received by the operation information reception unit, and the operation information for each user stored in the operation information storage unit!
- a first grouping unit that groups the users into a first group; a second grouping unit that groups the users into a second group that includes the first group grouped by the first grouping unit; Aggregation results of the operation information of users belonging to the first group grouped by the first grouping unit and the operation information of users belonging to the second group grouped by the second grouping unit
- a recommendation information generating unit that generates the recommendation information specific to the first group based on
- a recommendation information transmission unit that transmits the recommendation information generated by the recommendation information generation unit to the program viewing terminal, and the program viewing terminal receives the recommendation information transmitted by the recommendation information transmission unit.
- a program control unit that performs at least one of selection, recording, reproduction, and deletion of a program based on the recommendation information received by the recommendation information receiving unit.
- a program viewing terminal provides a program receiving unit that receives a program included in a broadcast wave or communication data, and the program received by the program receiving unit.
- An operation receiving unit that receives an input of the operation information representing a user's operation, and the operation information received by the operation receiving unit are connected to be communicable via a network and relate to a program that matches the user's preference
- An operation information transmitting unit that transmits recommendation information to a program recommendation server created according to the operation information, a recommendation information receiving unit that receives the recommendation information transmitted by the program recommendation server, and the recommendation information receiving unit
- a program control unit that performs at least one of selection, recording, reproduction, and deletion of a program based on the recommendation information received by the user.
- a program viewing program includes a program receiving unit that receives a program included in a broadcast wave or communication data, and a user operation on the program received by the program receiving unit.
- the operation reception unit that receives the input of the operation information representing the information, and the operation information received by the operation reception unit are connected to be communicable via a network, and the recommended information regarding the program that matches the user's preference.
- the operation information transmitting unit that transmits to the program recommendation server created according to the operation information, the recommendation information receiving unit that receives the recommendation information transmitted by the program recommendation server, and the recommendation information receiving unit. Based on the recommendation information, the computer is caused to function as a program control unit that performs at least one of program selection, recording, reproduction, and deletion.
- a program viewing method includes a program receiving step for receiving a program included in a broadcast wave or communication data, and for the program received in the program receiving step! /
- the operation accepting step for accepting input of the operation information representing the user's operation and the operation information accepted in the operation accepting step are communicably connected via a network and meet the user's preference.
- An operation information transmission step for transmitting recommended information related to a program to a program recommendation server created according to the operation information, a recommendation information reception step for receiving the recommendation information transmitted by the program recommendation server, and the recommendation information reception Based on the recommendation information received in the step, the program system performs at least one of program selection, recording, reproduction, and deletion. Including steps.
- a program recommendation server is connected to each other via a network so as to be communicable, and the program transmitted by a program viewing terminal for viewing the received program.
- An operation information receiving unit that receives operation information representing a user's operation, an operation information storage unit that stores the operation information received by the operation information receiving unit for each user, and an operation information receiving unit.
- a first grouping unit that groups the users into a first group based on the received operation information and the operation information for each user stored in the operation information storage unit;
- a second grouping unit that groups the users into a second group that includes the first group grouped by the first grouping unit, and the first group that is grouped by the first grouping unit.
- the recommendation information generation unit that generates the recommendation information specific to the first group based on the total result of the operation information of the user to which the user belongs, and the recommendation information generated by the recommendation information generation unit A recommendation information transmission unit for transmitting to the program viewing terminal.
- a program recommendation program is an operation representing a user operation on the program transmitted by a program viewing terminal that is connected to be communicable with each other via a network and views a received program.
- An operation information receiving unit that receives information; an operation information storage unit that stores the operation information received by the operation information receiving unit for each user; the operation information received by the operation information receiving unit; Based on the operation information for each user stored in the operation information storage unit, a first grouping unit for grouping the users into a first group, and the first grouped by the first grouping unit A second grouping unit for grouping the users into a second group including the first group, and the first grouping.
- a recommendation information generation unit that generates the recommendation information specific to the first group, and a recommendation information transmission unit that transmits the recommendation information generated by the recommendation information generation unit to the program viewing terminal. Make your computer work.
- the program recommendation method can communicate with each other via a network.
- the operation information receiving step for receiving operation information representing the user's operation on the program transmitted by the program viewing terminal connected to the program viewing terminal for viewing the received program, and the operation information received in the operation information receiving step.
- the operation information storage step stored in the operation information storage unit for each user, the operation information received in the operation information reception step, and the user stored in the operation information storage unit
- a second grooving step for grouping the users into a second group and a first grouping step.
- FIG. 1 is a conceptual diagram showing a recommended program extraction method according to the first embodiment.
- FIG. 2 is a block diagram showing the overall configuration of the program recommendation system of the first embodiment.
- FIG. 3 is a block diagram showing a configuration of a program viewing terminal of the first embodiment.
- FIG. 4 is a diagram showing a configuration of a program recommendation server according to the first embodiment.
- FIG. 5 is a flowchart showing a flow of processing up to transmission of reservation information in the program viewing terminal according to the first embodiment.
- FIG. 6 is a flowchart showing the flow of processing in the program recommendation server according to the first embodiment.
- FIG. 7 is a flowchart showing a flow of processing of power by receiving recommendation information in the program viewing terminal according to the first embodiment.
- FIG. 8 shows an example of the data structure of information stored in the reservation information DB of the program recommendation server. It is a figure which shows an example, (B) is a figure which shows the result which grouped the user who has the reservation information of program A, (C) is the grouping which grouped all the users who reserved program A.
- FIG. 5 is a diagram showing a reservation rate of programs other than program A in the program.
- FIG. 9 is a timing chart for explaining the marking method and marking counting method of the second embodiment.
- FIG. 10 is a graph showing the marking frequency of the second embodiment.
- Sono 11 is a block diagram showing the configuration of the program listening terminal of the second embodiment.
- Sono 12 is a block diagram showing the configuration of the program recommendation server of the second embodiment.
- FIG. 16 is a diagram illustrating an example of a data structure of marking information according to the second embodiment.
- FIG. 17 An example of the data structure of the marking DB of the second embodiment.
- FIG. 18 A timing chart showing the marking information of the second embodiment on a time axis.
- FIG. 19 An example of the data structure of the marking DB of the first application example of the second embodiment.
- FIG. 20 It is a conceptual diagram for explaining a marking information aggregation method of the first application example of the second embodiment.
- FIG. 21 A diagram showing an example of the data structure of the marking DB of the second application example of the second embodiment.
- FIG. 22 is a conceptual diagram for explaining a marking information counting method according to a second application example of the second embodiment.
- FIG. 26 is a flowchart showing a flow of processing until user cluster information is transmitted in the program viewing terminal according to the third embodiment.
- FIG. 27 is a flowchart showing the flow of processing in the program recommendation server according to the third embodiment.
- FIG. 28 is a first flowchart showing a flow of processing of power by receiving recommendation information in the program viewing terminal according to the third embodiment.
- FIG. 29 is a second flowchart showing the flow of power processing by receiving recommendation information in the program viewing terminal according to the third embodiment.
- FIG. 30 is a flowchart showing a process flow of a community reservation information statistics unit according to the fourth embodiment.
- FIG. 31 is a flowchart showing the flow of processing of an all-user reservation information statistics unit according to the fourth embodiment.
- FIG. 32 is a flowchart showing a process flow of the community specializing unit according to the fourth embodiment.
- FIG. 33 is a diagram for explaining extraction of recommended programs using tf′idf values.
- FIG. 34 shows the reservation status of users ⁇ to ⁇ belonging to community groups cl to c4 for program ⁇ .
- FIG. 1 is a conceptual diagram showing a recommended program extraction method according to the first embodiment of the present invention.
- a recording reservation operation will be described as a user's program operation.
- Graph 100a in Fig. 1 shows the reservation rate ranking of programs for which all users belonging to group A who have made a reservation are grouped after the user who has reserved "Drama A" is grouped. It is. That is, other reservations for users who have booked "Drama A” It is a program ranking. Therefore, if “Drama A” is reserved in common, V, the matching of preference conditions, collaborative filtering of other users ’preferences, and other reservation rate rankings other than“ Drama A ” It is the power to provide high-ranking programs as recommended programs to users.
- the group B including the users of the group A that is not only the group A of the users who reserved the “drama A” is extracted, and the reservation rate ranking of the group B is generated.
- Group B can be any grouping that includes all users belonging to group A! /, But all users who use the program recommendation system are group B to reduce processing. Is grooved as
- the reservation rate ranking of group A who reserved "Drama A" (graph 100a in Fig. 1) and the reservation rate ranking of group B consisting of all users (graph 100b in Fig. 1). ) are generated.
- the reservation rate ranking specialized for group A is generated by subtracting the reservation rate ranking of group B from the reservation rate ranking of group A (graph 100c in Fig. 1).
- each reservation rate ranking is determined by the reservation rate that is not the number of program reservations. Since the number of users belonging to Group A and Group B is different, the number of reservations is normalized by the number of users belonging to each group. This is because the processing is equivalent to the processing.
- the reservation rate ranking specialized for group A is determined as a threshold by threshold 105, and threshold 10 A program exceeding 5 is recommended as a recommended program 103 to the user. Therefore, this recommended program 103 can be provided as a recommended program specialized for group A.
- FIG. 2 is a block diagram showing the overall configuration of the program recommendation system according to the first embodiment.
- the program recommendation system includes a plurality of program viewing terminals 2 such as a program recording device and a program recommendation server 3.
- the plurality of program viewing terminals 2 and the program recommendation server 3 are communicably connected to each other via a general-purpose network 4 such as the Internet.
- a general-purpose network 4 such as the Internet.
- a plurality of program viewing terminals 2 are connected to the program recommendation server 3. 1S
- one program viewing terminal 2 is connected to simplify the explanation. The case where this is done will be described.
- the program viewing terminal 2 receives a program from a broadcast wave or an IP network, views the program, records the program, plays the recorded program, deletes the recorded program, Make a recording reservation.
- the program recommendation server 3 is connected to the program viewing terminal 2 and the network 4 so as to communicate with each other.
- the program recommendation server 3 acquires operation information representing the user's operation on the program from the program viewing terminal 2, and the program that matches the user's preference.
- the recommendation information about is created according to the operation information.
- FIG. 3 is a block diagram showing the configuration of the program viewing terminal of the first embodiment.
- the program viewing terminal 2 includes an antenna 201, a tuner 202, a TS (transport 'stream) decoder 203, a CAS (Conditional Access System) 204, a signal number canceling unit 205, a demnoplexer 206, an MPEG video decoder 207, an MPEG Audio decoder 208, HDMI (High-Definition Multimedia Interface) terminal 209, recording encryption unit 213, recording control unit 214, recording medium 215, playback control unit 216, recording decryption unit 217, PSI (Program Specific Information) Yuuki
- Tuner 202 receives a broadcast wave of a predetermined channel from broadcast waves such as terrestrial digital broadcast and BS digital broadcast, for example, and outputs the broadcast wave to TS decoder 203.
- the TS decoder 203 demodulates the received general transport stream stream and outputs it to the descrambling unit 205.
- CAS 204 is a general Conditional Access System, which reads out encryption key information from an IC card in which user registration information and program encryption keys are recorded, and outputs the information to descrambling unit 205.
- the descrambling unit 205 decrypts the encrypted broadcast signal demodulated by the TS decoder 203 based on the encryption key from the CAS 204, decrypts the encrypted program information and decrypts it. Output to multiplexer 206.
- the demultiplexer 206 separates information multiplexed as program information.
- program information at least video information, audio information and program management information are multiplexed. After these are separated, video information is extracted to MPEG Video decoder 207, audio information is extracted to MPEG Audio decoder 208, and program management information is extracted to PSI. Output to part 218.
- the MPEG Video decoder 207 demodulates the video information from the demultiplexer 206, converts it into video data, and outputs it to the HDMI terminal 209.
- the MPEG Audio decoder 208 demodulates the audio information input from the demultiplexer 206, converts it into audio data, and outputs it to the HDMI terminal 209.
- the HDMI terminal 209 is connected to an external HDMI terminal 210 via an HDMI cable, and the HDMI terminal 210 is connected to a monitor 211 and a speaker 212.
- HDMI terminals 209 and 210 are general video information and audio information interfaces, and the program viewing terminal 2 and the monitor 211 and the speaker 212 mutually authenticate each other and share the encryption key. And encrypting video data and audio data with a common encryption key And output to the monitor 211 and the speaker 212.
- the monitor 211 is a general program video information display device, which decrypts the encrypted video data received at the HDMI terminal 210 and outputs the video data of the program.
- the speaker 212 is a general program audio information output device that decrypts encrypted audio data received at the HDMI terminal 210 and outputs program audio data.
- the recording encryption unit 213 performs a signal processing for recording the received program information on a recording medium.
- the recording encryption unit 213 encrypts the program information input from the demultiplexer 206 and outputs the encrypted program information to the recording control unit 214. Therefore, the recording encryption unit 213 performs encryption processing specific to the recording medium for recording the program information.
- DVD-RAM uses CP RM (Content Protection for Rewritable Media) and Blu-ray ice cream HD-DVD uses AACS (Advanced Access Content System).
- the recording control unit 214 controls the recording of the program information encrypted by the recording encryption unit 213 on the recording medium.
- the recording control unit 214 records program information on the recording medium by performing modulation processing and error correction coding processing specific to the recording medium.
- the recording medium 215 is a computer-readable recording medium for recording program information.
- a magnetic recording disk such as a hard disk, an optical disk such as a DVD or a Blu-ray disk, or a semiconductor memory such as an SD card.
- magneto-optical recording media such as MO / MD are used.
- a program viewing program is recorded in the recording medium 215, and the program viewing program is read and executed by a CPU (Central Processing Unit) or the like.
- the reproduction control unit 216 reads program information recorded on the recording medium 215.
- the recording control unit 214, the recording medium 215, and the reproduction control unit 216 are generally configured as a drive system.
- the recording decryption unit 217 performs decryption corresponding to the encryption of the recording encryption unit 213.
- the recording decryption unit 217 decrypts the program information encrypted and recorded on the recording medium 215 and outputs the decrypted program information to the TS decoder 203. Note that the reproduction process of the program information output to the TS decoder 203 is the same as the reproduction process of the program received by the antenna 201 described above.
- the PSI extraction unit 218 uses the program management information extracted by the demultiplexer 206 to generate PSI (Prog ram Specific Information) is extracted.
- PSI Prog ram Specific Information
- PMT Program
- Map Table Map Table
- EIT Event Information Table
- TOT Time Offset Table
- the PMT stores packet identification information such as images and sounds included in the program. Therefore, if packet identification information such as images and sounds can be obtained from the PMT, the program can be played by extracting the packets with the packet identification information.
- the EIT includes information related to the program such as the program name, broadcast date and time, and broadcast content.
- EPG Electronic Program Guide
- the EPG storage unit 219 creates and stores EPG information using the EIT in the PSI extracted by the PSI extraction unit 218.
- Key 220 accepts input from the user and is configured with a remote control or a front panel.
- Reservation management unit 221 accepts a program to be reserved for recording in accordance with a command input with key 220, and creates reserved program information.
- the reserved program information includes program identification information for identifying a program.
- the reserved program information is created using the £ 0 information stored in the EPG storage unit 219.
- the reserved memory 222 includes a memory that stores reserved program information created by the reservation management unit 221.
- the reserved program information may include program broadcast date and time, channel information, and broadcast station identification information instead of program identification information. That is, the program can be identified by the broadcast date and time of the program, channel information, and broadcast station identification information.
- the user identification unit 223 identifies a user who uses the program viewing terminal 2.
- the user identifying unit 223 identifies a user who uses the program viewing terminal 2 based on user information input using the key 220.
- the user information DB 224 is a database in which user information of all users who use the program viewing terminal 2 is registered.
- the user identification unit 223 identifies a user based on user information input using the key 220 and user information stored in the user information DB 224 in advance. Further, when the user identification is successful, the user identification unit 223 outputs the user information of the user currently in use to the reservation information transmission unit 225.
- the reservation information transmission unit 225 is a general network interface, and is a general-purpose network interface. After establishing a network with the program recommendation server 3 connected via the network 4, the program recommends the reservation information including the reserved program information stored in the reservation memory 222 and the user information input from the user identification unit 223. Send to server 3.
- the reservation memory 222 stores the reserved user in association with the user identification unit 223 and stores the reserved program information that has been transmitted to the program recommendation server 3 by the reservation information transmission unit 225 and the untransmitted information. Manage reserved program information in an identifiable manner.
- the recommendation receiving unit 226 receives recommendation information regarding a program recommended by the program recommendation server 3 via the general-purpose network 4.
- the recommended program information extracting unit 227 extracts recommended program information from the recommended information received by the recommendation receiving unit 226.
- Recommended program information is information that can uniquely identify a recommended program. For example, program identification information for identifying a program, or broadcast station identification for identifying a program's broadcast date and time, channel information, and a broadcast station. Contains information.
- the extracted recommended program information is output to the program reservation determination unit 228.
- the recommended information includes recommended program information and popular program information.
- the recommended program information is information representing a program specialized in the preference of the user who uses the program viewing terminal 2
- the popular program information is information representing a program popular with all users.
- the program reservation determination unit 228 determines whether the recommended program indicated by the recommended program information is a program broadcast in the past. That is, the program reservation determination unit 228 determines that the program has been broadcast in the past if the broadcast date / time of the recommended program is earlier than the current date / time. When it is determined that the recommended program is not a program broadcasted in the past !, the program reservation determination unit 228 receives the reserved program information stored in the reservation memory 222 and the recommended program information extraction unit 227. Based on the recommended program information, whether or not the corresponding program is already reserved is determined.
- the program reservation determination unit 228 outputs the recommended program information to the reservation management unit 221 to perform reservation processing corresponding to the recommended program information. . If it is determined that the program is reserved, the program reservation determination unit 228 notifies the user that the recommended program has already been reserved, and ends the process. Further, when it is determined that the recommended program indicated by the recommended program information is a program broadcast in the past, the program reservation determining unit 228 outputs the recommended program information to the recommended program extracting unit 229. The recommended program extracting unit 229 determines whether or not the program specified by the recommended program information is recorded on the recording medium 215.
- the recommended program extraction unit 229 notifies the user that the recommended program is not recorded, and ends the process.
- the recommended program extracting unit 229 notifies the user that the recommended program exists, and prompts the user to play it.
- the playback control unit 216 calls the program information indicated in the recommended program information from the recording medium 215 and plays back the program.
- the recommended program extraction unit 229 determines whether the program specified by the recommended program information is recorded in the recording medium 215! /, Or not, in the recommended program information from the program recommendation server 3 Judgment is based on the program identification information of the program, or the broadcast date and time of the recommended program, channel information, and broadcast station identification information.
- FIG. 4 is a diagram illustrating a configuration of the program recommendation server according to the first embodiment.
- the program recommendation server 3 includes a reservation information reception unit 230, a user management unit 231, a user attribute DB (database) 232, a reservation information registration unit 234, a reservation information DB (database) 235, a specialization generation unit 30, and a recommended program It consists of a creation unit 239 and a recommendation transmission unit 240.
- the specialization generation unit 30 of the program recommendation server 3 includes a community reservation information statistics unit 236, an all-user reservation information statistics unit 237, and a community specialization unit 238.
- the reservation information receiving unit 230 receives the reserved program information of the program reserved by each user from the reservation information transmitting unit 225 of the plurality of program viewing terminals 2 connected to the general-purpose network 4, and the user who made the reservation Reservation information including user information for identifying.
- the reserved program information includes program identification information that can uniquely identify the reserved program, or program broadcast date and time, channel information, and broadcast station identification information of the reserved program.
- the received reservation information is output to the user management unit 231.
- the user attribute DB 232 includes the user information of the registered user, and, if necessary, the user name, gender, address / phone number, manufacturer / model of the program viewing terminal, the resolution of the monitor to be connected, and the network of the program viewing terminal.
- User attribute information such as IDs is managed as a user registration list.
- a user registers in advance with the program recommendation server 3 for user recommendation. Perform the recording process.
- the program recommendation server 3 After user registration information such as user name setting and address / telephone number notification is given, the program recommendation server 3 generates user information and notifies the user on the program recommendation server 3 side.
- the user management unit 231 determines whether or not the user is a registered user of the user power program recommendation server 3 specified by the received user information.
- the user management unit 231 searches the user registration list stored in the user attribute DB 232 for user information that matches the received user information. If it is determined that the received user information is registered in the user attribute DB 232, it is determined that the user information is a registered user of the program recommendation server 3, and the processing is continued.
- the registered user is registered in the program viewing terminal 2 used by the user as an unregistered user in the program recommendation server 3 Sends an error message to the effect and interrupts further processing.
- the program viewing terminal 2 that has received the error message displays the error message on the monitor 211.
- the reservation information registration unit 234 displays the reserved program information for each user received by the reservation information reception unit 230. It is stored in the reservation information DB 235 in association with the received user information.
- the reserved information DB 235 is a database that stores reserved program information transmitted by the program viewing terminal 2 in association with user information.
- the community reservation information statistics unit 236 groups all users having the same reserved program information as the reserved program information as a community group based on the reserved program information received by the reservation information receiving unit 230. To do. In addition, the community reservation information statistics unit 236 generates rankings by counting reservation program information other than the reserved program information of all users belonging to the community group for each program, and the total number of users belonging to the community group. Normalize the ranking and calculate the reservation rate ranking for each program in the community group. The community reservation information statistics unit 236 outputs the normalized ranking (reservation rate ranking) to the community specialization unit 238.
- the all user reservation information statistics unit 237 converts all users having reservation program information received by the reservation information reception unit 230 into a community gnole. All registered users, that is, all registered users of the program recommendation server 3 are grouped as all user groups including community groups. In addition, the all-user reservation information statistics unit 237 performs a statistical process on the reservation program information of all users after grouping all users, and creates a reservation rate ranking for each program. In other words, the all-user reservation information statistics unit 237 generates the ranking by summing up the reserved program information of all registered users for each program, normalizes the ranking by the number of all registered users, and creates a ranking within the group formed by all registered users. The reservation rate ranking of each program is calculated. The created reservation rate ranking is output to the community specialization unit 238 and the recommended program creation unit 239.
- the community specialization unit 238 subtracts the reservation rate ranking calculated by the all-user reservation information statistical unit 237 from the reservation rate ranking calculated by the community reservation information statistical unit 236 to obtain community reservation information.
- the reservation rate ranking specialized for the community group grouped by the statistics unit 236 is created and output to the recommended program creation unit 239.
- Ranking subtraction subtracts the reservation rate for each program for each program.
- the recommended program creation unit 239 Based on the reservation rate ranking generated by the community specialization unit 238, the recommended program creation unit 239 extracts a program having a reservation rate higher than a preset threshold as a recommended program, and selects the recommended program.
- Program identification information for identification, or recommended program information including broadcast date and time of the recommended program, channel information on which the recommended program is broadcast, and broadcast station identification information for identifying a broadcast station on which the recommended program is broadcast Create Accordingly, the recommended program creating unit 239 creates recommended program information specialized for the community group.
- the reservation rate is higher than a preset threshold! /, And the ability to extract a program as a recommended program.
- the present invention is not particularly limited to this, and the community specializing unit 238 Therefore, a predetermined number of programs that are higher in the reservation rate ranking generated may be extracted as recommended programs.
- the recommended program creation unit 239 is generated by the all-user reservation information statistics unit 237. Based on the reservation rate ranking, the reservation rate is higher than a preset threshold! /, The program is extracted as a popular program, program identification information for identifying the popular program, or the broadcast date and time of the popular program, Popular program information including channel information for broadcasting a popular program and broadcasting station identification information for identifying a broadcasting station for broadcasting the popular program is created.
- the reservation rate is higher than a preset threshold! /, The ability to extract programs as popular programs.
- the present invention is not particularly limited to this, and the all-user reservation information statistics unit The predetermined number of programs with the highest reservation rate ranking generated by 237 may be extracted as popular programs.
- the recommended transmission unit 240 transmits the recommended information including the recommended program information and popular program information created by the recommended program creating unit 239 to the program viewing terminal 2.
- the program recommendation server 3 includes a reproduction control unit (not shown) that reads a program and data from a computer-readable recording medium.
- a recording medium for example, a magnetic recording disk such as a hard disk, an optical disk such as a DVD or a Blu-ray disk, a semiconductor memory such as an SD card, and a magneto-optical recording medium such as MO / MD are used.
- a program recommendation program is recorded on this recording medium, and is read and executed by a program recommendation program force SCPU (central processing unit) or the like.
- the user A reserves the program A by the program viewing terminal 2 owned by the user A.
- the program viewing terminal 2 transmits user information for identifying the user A and reserved program information for specifying the program A to the program recommendation server 3.
- the reservation information receiving unit 230 of the program recommendation server 3 receives user information for identifying the user A and reserved program information for specifying the program A from the program viewing terminal 2 owned by the user A.
- the user management unit 231 determines whether or not the received user information is stored in the user attribute DB 232, and if stored, determines that the user A is a registered user of the program recommendation server 3 and continues the process.
- the reserved information registration unit 234 registers reserved program information for specifying the program A in the reserved information DB 235 in association with user information for identifying the user A.
- the community reservation information statistics unit 236 is associated with reserved program information for identifying program A! /, All users are extracted from the reservation information DB 235 to form a community group.
- the community reservation information statistics unit 236 statistically processes programs other than program A in the community group. Then, the community reservation information statistics unit 236 calculates the reservation ratio of programs other than program A in the community group, generates a reservation ratio ranking in which programs are arranged in descending order of the reservation ratio, and generates the generated reservation ratio ranking. Is output to Community Specialization Department 238.
- the all-user reservation information statistics unit 237 includes the community groups formed by extracting all users registered in the user attribute DB 232 of the program recommendation server 3 from the reservation information DB 235 and forming the community reservation information statistics unit 236. Form all user groups.
- the all-user reservation information statistics unit 237 extracts reserved program information associated with user information in all user groups from the reservation information DB 235, and statistically processes programs other than program A. Then, the all-user reservation information statistics unit 237 calculates the reservation ratio of programs other than program A in all user groups, generates a reservation ratio ranking in which the programs are arranged in descending order of the reservation ratio, and generates the generated reservation ratio. The ranking is output to the community specialization unit 238 and the recommended program creation unit 2 39.
- the community specialization unit 238 subtracts the reservation rate ranking of the program created by the all-user reservation information statistics unit 237 from the program reservation rate ranking created by the community reservation information statistics unit 236. As a result, the community specializing unit 238 creates a program reservation rate ranking specialized for the community group of the user who has reserved the program A, and outputs it to the recommended program creating unit 239.
- the recommended program creation unit 239 Based on the reservation rate ranking specialized for the community group created by the community specialization unit 238, the recommended program creation unit 239 has a reservation rate higher than the threshold! /, Program reservation program information or ranking.
- the recommended program information is created by extracting the top reserved program information.
- the recommended program creation unit 239 is based on the generalized reservation rate ranking of all registered users created by the all-user reservation information statistics unit 237, and the reserved program information or ranking of programs with a reservation rate higher than the threshold value. Top program information is extracted and popular program information is created.
- the recommendation transmitting unit 240 outputs the recommended program information and popular program information created by the recommended program creating unit 239 to the program viewing terminal 2 of the user A. Next, the process flow of the program recommendation system will be further described. FIGS.
- FIG. 5 to 7 are flowcharts showing the processing flow of the program recommendation system according to the first embodiment, and FIG. 5 shows the processing until the reservation information is transmitted in the program viewing terminal according to the first embodiment.
- 6 is a flowchart showing a flow of processing in the program recommendation server according to the first embodiment, and FIG. 7 is recommended information in the program viewing terminal according to the first embodiment. It is a flowchart which shows the flow of processing after receiving
- key 220 accepts input of user information by the user and accepts login to program viewing terminal 2 (step Sl).
- the user identification unit 223 determines whether or not the user authentication is successful, that is, whether or not the input user information matches the user information registered in advance in the user information DB 224 (step S2 ).
- the user identification unit 223 outputs an error notification to the monitor 211 and ends the processing (step S3 ).
- tuner 202 receives a broadcast wave from antenna 201 (step S4).
- TS decoder 203 demodulates the broadcast wave received by tuner 202.
- the signature canceling unit 205 uses the decryption key from the CAS 204 to decrypt the program information contained in the broadcast wave.
- the demultiplexer 206 separates the multiplexed program information into a video signal, an audio signal, and a program control signal.
- the PSI extraction unit 218 extracts the PSI included in the separated program control signal (step S5).
- PSI extraction section 218 creates EPGs of all receivable channels based on the extracted PSI and stores them in EPG storage section 219 (step S6).
- the PSI extraction unit 218 displays the EPG created on the monitor 211 so that individual programs can be selected from the user (step S7).
- the reservation management unit 221 determines whether or not there has been a reservation operation from the user (step S8).
- Reservation management unit 221 can be selected by the user from the EPG that displays the program as selectable. Accept pre-reservation operations and accept them. .
- pre-reservation The management management unit 222211 stands by while the pre-reservation operation by the user is being performed. .
- Pre-reservation contract management management department 222211 is the program program identification identification information corresponding to the program group that has become the target of the pre-reservation operation.
- the broadcast date of the program, or the program program broadcast date the information on the information on the channel information and the information on the broadcasting station identification information is extracted from the EEPPGG.
- the pre-reservation memory 222222 the pre-reservation number program group information is recorded and stored. .
- the pre-reservation contract management management unit 222211 sets the pre-reservation contract for the program group (step SS99). .
- program program group AA the program program group that has been set for the advance reservation in step SS99 is called program program group AA. I will explain it. .
- the pre-reservation agreement information information transmission / reception unit 222255 Distinguish between pre-reserved reservation program information information AA corresponding to program group AA and the user who made pre-reserved reservations for program group AA.
- the user information information for suru-takuta is sent as a pre-reservation information information to the recommended program group recommended server 33 ((step SS 1100)). .
- the pre-reservation reservation information report reception / reception unit 223300 of the program group recommendation recommendation server 33 is the program group viewing program. End of listening terminal
- the pre-reservation information information sent and received in accordance with 22 is received and received ((Step SSll)). .
- the user management administration department 223311 determines whether or not the user authentication certificate has succeeded, that is, the advance reservation received and received.
- User information included in the contract information information is registered in advance in the DDBB223322 attribute of the user information attribute of the program group recommendation recommendation server 33 Judgment is made as to whether or not it agrees with the information (Step Step SS 1122). .
- the user authentication certificate fails and fails, and it is judged that the user is not a registered user ((NNOO in step SS1122))
- the user management department 223311 creates an error message to indicate the fact and displays the program program viewing / listening terminal.
- An error message message is sent to the terminal end 22 to send and receive it, and the processing is terminated (step SS1133).
- the terminal program viewing / listening terminal terminal 22 that received and received the error message message failed and failed in the user authentication certificate.
- the pre-reservation contract information report registration / recording unit 223344 is the user information received / received by the pre-reservation contract information report reception / reception unit 223300.
- the reserved reservation program information information AA in the reserved reservation information information DDBB223355 corresponds to the information information and registering the reserved reservation program information information AA in the reserved reservation information information DDBB223355 ((Step Step SS 1144)) . .
- the pre-reservation contract information information registration / registration section 223344 is used by all the registered registration users stored in the pre-reservation contract information information information DDBB223355. Based on the scheduled reservation program information information, all the users who have the reserved program number information AA of the reserved program group AA are extracted, , Komi Muninitte
- the communimunity pre-reservation contract information information statistics section 223366 is connected to the communimunity GN Relay
- the reserved program information of programs other than the program A associated with the user information of the user of the group is totaled to generate a ranking of the number of reservations for each program (step S16).
- the community reservation information statistics unit 236 normalizes the total number of reservations for each program by the number of grouped users, generates a reservation rate for each program, and creates a program with a high reservation rate.
- the reservation rate rankings arranged in order are generated (step S17).
- the all-user reservation information statistics unit 237 extracts users registered in the user attribute DB 232, that is, all registered users of the program recommendation server 3, and groups them as all user groups (step S). 18). Since all the user groups created in step S18 are composed of all registered users, the group including the user group having the reserved program information of program A created in step S15 is included. Become.
- the all-user reservation information statistics unit 237 aggregates the reserved program information of all users of programs other than the program A in the grouped all user groups, and ranks the number of reservations for each program. Generate (step S19).
- the all-user reservation information statistics unit 237 normalizes the total number of reservations for each program by the number of grouped users, that is, the total number of users registered in the program recommendation server 3. Then, the reservation rate for each program is generated, and the reservation rate ranking arranged in order from the program with the highest reservation rate is generated (step S20).
- the community specializing unit 238 determines the total user reservation information statistics unit 237 from the reservation rate ranking of programs other than the program A of the user who has reserved the program A generated by the community reservation information statistics unit 236.
- the reservation rate ranking specialized for the user who reserved the program A is generated by subtracting, for each program, the reservation rate ranking of programs other than the program A generated by all users.
- the recommended program creation unit 239 creates recommended program information by extracting programs with a reservation rate higher than a predetermined threshold from the reservation rate rankings generated by the community specialization unit 238 (step S21). Note that the recommended program creation unit 239 may extract recommended programs by extracting a predetermined number of programs from the top programs in the reservation rate ranking generated by the community specialization unit 238.
- the recommended program creation unit 239 selects a number having a reservation rate higher than a predetermined threshold from among the reservation rate rankings of all users generated by the all user reservation information statistics unit 237.
- Popular program information is created by extracting sets (step S22). Note that the recommended program creation unit 239 may create popular program information by extracting a predetermined number of programs from the higher ranking program power among the reservation rate rankings generated by the community specialization unit 238.
- the recommended transmission unit 240 transmits the recommended program information created by the recommended program creating unit 239 and recommended information including popular program information to the program viewing terminal 2 (step S23).
- the recommendation receiving unit 226 of the program viewing terminal 2 receives the recommended information including the recommended program information and the popular program information transmitted by the recommended transmitting unit 240 of the program recommendation server 3 ( Step S31).
- the recommended program information extraction unit 227 determines whether or not the recommended information received by the recommendation receiving unit 226 is also the recommended program information extracted (step S32).
- the recommended program information extraction unit 227 prompts the user to select one of recommended program information and popular program information.
- the recommended program information is extracted when the recommended program information is selected, and the popular program information is extracted when the popular program information is selected. It is also possible to provide a setting to extract both recommended program information and popular program information. This can provide more options to the user.
- the program reservation determination unit 228 includes program identification information included in the recommended program information extracted by the recommended program information extraction unit 227, Alternatively, using the program broadcast date / time, channel information, and broadcast station identification information, it is determined whether or not the reservation has already been completed by the program viewing terminal 2 (step S33). For example, the program reservation determination unit 228 determines whether or not the program identification information included in the recommended program information matches the program identification information included in the reserved program information stored in the reservation memory 222.
- the reservation management unit 221 performs reservation setting for the recommended program (step S34). After the recommended program reservation setting is completed, or when the recommended program reservation setting has been completed in advance (YES in step S33), the program reservation determination unit 228 reserves all recommended program information included in the recommended information. It is determined whether or not the setting is completed (step S35).
- step S35 When it is determined that reservation of all recommended program information has been completed! /, NA! / (N in step S35) O) Return to step S33 and repeat the reservation setting process for the remaining recommended program information. If it is determined that the reservation of all recommended program information has been completed (YES in step S35), the process ends.
- program reservation determination unit 228 extracts recommended program information. Using the program identification information included in the popular program information extracted by the section 227 or the program broadcast date / time, channel information, and broadcast station identification information, the program viewing terminal 2 determines whether or not the reservation has already been completed. (Step S36). For example, the program reservation determination unit 228 determines whether or not the program identification information included in the popular program information matches the program identification information included in the reserved program information stored in the reservation memory 222. .
- the reservation management unit 221 makes a reservation setting for the popular program (step S37).
- the program reservation determination unit 228 reserves all the popular program information included in the recommended information. It is determined whether or not the setting is completed (step S38).
- step S38 If it is determined that reservation of all popular program information has been completed! /, NA! / (NO in step S38), the process returns to step S36 to perform reservation setting processing for the remaining popular program information. Execute repeatedly. If it is determined that all popular program information has been reserved (YES in step S38), the process is terminated.
- the recommended program extraction unit 229 of the program viewing terminal 2 determines whether or not the recommended program has been recorded on the recording medium 215. If the program has been recorded, the recommended program extraction unit 229 may output a message to the monitor 211 prompting the user to play.
- Digital recorders in recent years include those that automatically record programs by EPG keyword search and those that continue to record all channels for 24 hours. Therefore, it is easy to consider that the recommended program is recorded even if the user does not intentionally set the reservation or record. Therefore, a system that prompts the user to play back a program that has missed viewing can be provided by the same system.
- the program recording reservation operation has been described as the user operation.
- the present invention is not limited to this.
- the user operation may be any operation as long as the user is an operation related to the preference for the program.
- the user operation may be a program playback operation, a recording operation, or an erase operation.
- the program viewing terminal 2 transmits the reproduced program identification information to the program recommendation server 3.
- the reservation information registration unit 234 of the program recommendation server 3 manages the programs played back for each user, and the community reservation information statistics unit 236 groups the users who played the same program and plays them back by the users in the group. Generate a ranking for the playback rate of the program. Further, the all-user reservation information statistics unit 237 groups all registered users and generates a playback rate ranking for all registered users.
- the community specialization unit 238 subtracts the reproduction rate ranking of the program created by the all-user reservation information statistics unit 237 from the reproduction rate ranking of the program created by the community reservation information statistics unit 236. Thus, the community specialization unit 238 generates a reproduction rate ranking specialized for the user who has played the specific program, and the recommended program creation unit 239 is based on the reproduction rate ranking generated by the community specialization unit 238. To generate recommended program information.
- the program viewing terminal 2 transmits identification information of the recorded program to the program recommendation server 3.
- the reservation information registration unit 234 of the program recommendation server 3 manages the programs recorded for each user, and the community reservation information statistics unit 236 groups the users who recorded the same program, and records the users in the group. Generate program recording rate ranking.
- the All User Reservation Information Statistics Unit 237 groups all registered users. Then, the recording rate ranking of all registered users is generated.
- the community specialization unit 238 subtracts the recording rate ranking of the program created by the all-user reservation information statistics unit 237 from the recording rate ranking of the program created by the community reservation information statistics unit 236.
- the community specializing unit 238 generates a recording rate ranking specialized for the user who recorded the specific program, and the recommended program creating unit 239 is based on the recording rate ranking generated by the community specializing unit 238. Generate recommended program information.
- the present invention can also be applied when the user operation is a program erasing operation.
- the program viewing terminal 2 transmits the erased program identification information to the program recommendation server 3.
- the reservation information registration unit 234 of the program recommendation server 3 manages the deleted programs for each user, and the community reservation information statistics unit 236 groups the users who deleted the same program and deletes the users in the group. Generate program erasure rate ranking.
- the all-user reservation information statistics unit 237 creates a deletion rate ranking for all registered users by grooving all registered users.
- the community specialization unit 238 subtracts the deletion rate ranking of programs created by the all-user reservation information statistics unit 237 from the deletion rate ranking of programs created by the community reservation information statistics unit 236.
- the community specializing unit 238 generates a deletion rate ranking specialized for the user who deleted the specific program, and the recommended program creation unit 239 recommends based on the deletion rate ranking generated by the community specializing unit 238.
- Program information is generated.
- the program information that has been replayed can be aggregated to promote a reservation operation for programs to be broadcast in the future. At this time, it is possible to create recommended information for prompting playback by grouping users having the same reservation information and counting the programs played back in the group. Conversely, if the program information is grouped and the reservation information is aggregated within the group, it is based on the reservation information of users who have similar preferences playing the same program! It is also possible to automatically perform a program reservation operation that matches the preferences.
- Different operations may be combined as user operations.
- a ranking may be created by totaling together a program accompanied by a reproduction operation and a program accompanied by a recording operation.
- the weighting for aggregation is based on different operations. It may be changed. For example, it is determined that the user's preference is higher for a reproduced program than for a program that has been recorded but not reproduced. Therefore, a ranking is created with weighting based on this. For example, 1 point is recorded when only a specific program is recorded, 2 points when it is being played, and 1 point when there is an erase operation. This makes it possible to generate a more accurate ranking for user preferences.
- preference operation information for a program is aggregated by the program recommendation server 3 for each program viewing terminal 2 (for each user), and users with similar user preferences are grouped and duplicated.
- the recommended filtering is recommended to the user by collaborative filtering, and the recommended program is automatically recorded, urged to be played, deleted, and deleted. Support program viewing.
- a second group that includes the first group is created by grouping users based on the program to be operated. Subtract the ranking of the second group from the ranking of the first group, i.e.
- FIG. 8 is a diagram illustrating a data structure of the reservation information DB 235 according to the first embodiment.
- FIG. 8A shows an example of a data structure of information stored in the reservation information DB 235 of the program recommendation server 3.
- reservation information indicating that user A has reserved program A is received by reservation information receiving section 230 of program recommendation server 3, and reservation information registration section 234 receives reservation information.
- the data structure of the reservation information DB 235 when the reserved program information 402 of the program A is registered in the list 401 of the user A of the DB 235 is shown.
- Reservation information DB235 is the user identification information for all registered users of program recommendation server 3 (user This is a data structure in which user identification information for managing (ID) is registered in the column direction, and reserved program information that the user has reserved for recording is registered in the row direction corresponding to each user identification information.
- users B, C, and D are registered. Users B, C, and D indicate registered users other than User A, and reserved program information corresponding to the users B, C, and D is registered in the row direction of each user.
- user B is associated with program A, program C, program D and program E, and reserved program information
- user C is referred to as program X, program Y, program ⁇ and program S.
- Reservation program information is associated
- user D is associated with program ⁇ , program ⁇ , program C and program ⁇ and V, and reservation program information!
- FIG. 8 ( ⁇ ) is a diagram showing a result of grouping users having reservation information of program ⁇ .
- the result of grouping users having reservation information of program A by the community reservation information statistics unit 236 is shown.
- all users other than user C have reservation information for program A! /, So users A, B, and D other than user C are grouped.
- FIG. 8C is a diagram showing a reservation rate of programs other than program A in a group in which all users who have reserved program A are grouped.
- the reservation frequency of programs other than program A is ranked by the community reservation information statistics unit 236 in a group composed of users who have reserved program A.
- program C is most frequently reserved in addition to program A, and then program B, program D, and program E are reserved.
- the program reservation rate ranking is normalized by the number of grouped users, that is, 3 people, and the program C reservation rate is calculated as 100%, and programs B, D, and E are calculated as 67%. .
- normalization may be performed by the total frequency of ranking.
- the total frequency is 9. Therefore, the probability of program C is normalized to 33%, and the probability of program B, program D, and program E is normalized to 22%.
- the reservation probability for each program in the group is obtained, so that an accurate ranking can be generated.
- a second embodiment of the present invention will be described.
- user operation information for each program is collected in the program recommendation server, and users are grouped based on user operations for the same program, and the user's favorite programs in the group are ranked.
- recommended programs specific to groups are extracted and recommended programs are provided to users.
- the recommended scenes and recommended time zones in a program are provided by aggregating user operations for a part specified on the time axis in the program rather than collecting user operation information for each program. To do.
- a user operation for marking a user's favorite scene in a program will be described as a user operation for a part specified on the time axis in the program.
- Marking is the marking of a favorite scene while the user is watching a broadcast program or a recorded program. For example, the user keeps pressing a predetermined favorite scene button between favorite scenes. Marking by entering the start time and end time of the favorite scene, marking to specify the duration of the favorite scene by operating the button at the start time of the favorite scene, and at the end time of the favorite scene For example, you can mark the duration of favorite scenes by operating buttons.
- the program viewing terminal for each user is marked with! /
- the program is marked for each user, and the marking information and the program identification information are transmitted to the program recommendation server.
- the program recommendation server groups users marking at the same time, users with similar temporal markings, or users with overlapping marking times, and aggregates marking information within the group.
- the recommended scene specialized in is generated and the recommended scene is provided to the user.
- FIG. 9 is a timing chart for explaining the marking method for the specific programs of user A and user B and the marking totaling method.
- a program stream 500 shown in FIG. 9 shows the time course of a broadcast program that progresses in the right direction in the drawing.
- the program stream 500 is composed of continuous program frames.
- user A is marking the specific program.
- User A marks a scene between time T1 and time ijT3 (first marking 501) and marks a scene between time T6 and time T8 (second time) in this specific program. Marking 502).
- User B marks the scene from time T2 to time T4 (first marking 503) and marks the scene from time T5 to time T7 in the specific program (second marking). 504) and marking the scene between time T9 and time T10 (third marking 505).
- the user A's program viewing terminal transmits the user A's user information, the program identification information of the specific program, and the user A's marking information to the program recommendation server.
- the marking information represents the time marked in a specific program, and includes, for example, the time when marking is started and the time when marking is finished.
- the marking start time T1 and marking end time IJT3 in the first marking 501 and the marking start time T6 and marking end time T8 in the second marking 502 are generated as marking information.
- the user B's program viewing terminal sends the user B's user information, the program identification information of the specific program, and the user B's marking information to the program recommendation server.
- marking start time ⁇ IJT2 and marking end time ⁇ ⁇ ⁇ ⁇ IJT4 in the first marking 503, marking start time T5 and marking end time T7 in the second marking 504, and the third marking 503 Marking 505 at the start of marking ⁇ IJT9 and marking end time T10 are generated as marking information.
- user B views the specific program earlier than user A and transmits the marking information. Therefore, when user A transmits the marking information, the marking information of user B is not displayed.
- the following explanation is given assuming that the program is already registered in the program recommendation server.
- the user information is given to the user from the program recommendation server by registering the user information in the program recommendation server, and is managed in the program viewing terminal used by the user. It is done.
- the program identification information is given for each program! /, The program identification information for identifying the program, and the identification indicated by the broadcast date / time of the program, channel information and broadcast station identification information. Information that can be misaligned.
- the marking information includes a marking start time and a marking end time.
- time T1 is included as the marking start time
- time T3 is included as the marking end time. If the marking time is specified by the marking start time and marking width by the user, the marking time is changed to the marking start time and the marking end time and transmitted.
- the program recommendation server When receiving the user A marking information from the user A program viewing terminal, the program recommendation server analyzes the received marking information and groups users having similar marking information. Note that users having similar marking information are users whose marking positions coincide in time and users who have overlapping marking positions. Also, when determining the overlap of marking positions, in order to accurately determine the identity of user preferences, it is determined whether there is an overlap of 50% or more over the entire marking time, and 50% or more Group users with overlapping.
- a first overlap 506 and a second overlap 507 indicate the overlap between the user A marking information and the user B marking information.
- the total time of the first overlap 506 and the second overlap 507 is 50 with respect to the total marking time of user A (the total time of user A's first marking 501 and second marking 502).
- User A and User B are grouped as the same group because they are more than%. In this embodiment, it is assumed that a group is composed of only user A and user B. However, all users with overlapping marking times of 50% or more are grouped into the same group.
- the users are grouped based on the user's marking operation on the program.
- Fig. 10 is a graph showing the marking frequencies of programs B and C in which similar marking information exists in programs other than the specific program of the grouped user. It is.
- the frequency distribution graph 600a in FIG. 10 groups users having marking information similar to the marking information of the specific program of user A, and displays similar marking information for programs other than the specific program.
- Program B and program C are extracted, and the frequency distribution indicates the overlap of marking information of users in the groups in program B and program C.
- the first peak 601 and the second peak 602 show high frequencies.
- Program C the first peak 603 and the second peak 604 show high frequencies. This is because in a group of users with marking information similar to the marking information for user A's specific program, program B first peak 601, program B second peak 6 02, program C first Show that many markings have been made at peak 603 and the second peak 604 of program C! /.
- the frequency distribution graph 600b in Fig. 10 shows programs of all registered users of the program recommendation server as groups including a group of users having marking information similar to the marking information of the specific program of user A. This is a frequency distribution showing the overlap of marking information in B and Program C.
- the frequency distribution graph 600b of all users a relatively large number of markings are performed at the first peak 605 and the second peak 606 in the program B.
- the marking frequency is low on the first peak 607 and the second peak 608 as a whole!
- the frequency distribution graph 600c in Fig. 10 shows the frequency obtained by normalizing the marking frequency of a group consisting of users with marking information similar to the marking information of user A with the total marking time of all users in the group.
- the distribution graph 600a shows the result of subtracting the frequency distribution graph 600b obtained by normalizing the marking frequency of all users by the total marking time of all users.
- the first peak 601 in the group in program B and the first peak 605 of all users have a time lag, so the first difference peak 609 of program B is generated. .
- the second peak 602 in the group of program B has a marking frequency almost equal to the second peak 606 of all users. 610 does not have a large marking power.
- the frequency distribution within the group shows that the first peak 603 and the second peak 604 have a high marking frequency, but the frequency distribution of all users Neither peak 607 of 1 nor peak 608 of 2 has a high frequency. Therefore, in the program C, the first difference peak 611 and the second difference peak 612 have high frequencies.
- the difference frequency distribution graph 600c in FIG. 10 performs threshold determination based on the marking frequency of the difference for each program and a predetermined threshold 620, and recommends scenes having a marking frequency higher than the threshold 620 as recommended scenes.
- program B the scene from time T1 to time T2 of program B is generated as recommended scene 1
- program C from time T3 to time T4 of program C is generated as recommended scene 2.
- the recommended scene 3 is extracted from time T5 to time T6 of program C.
- the recommendation degree for each program has priority over the program power having a large integrated value of the marking frequency on the time axis for each program of each recommended scene extracted in the frequency distribution of differences.
- FIG. 11 is a block diagram showing the configuration of the program viewing terminal of the second embodiment
- FIG. 12 is a block diagram showing the configuration of the program recommendation server of the second embodiment.
- the program recommendation system of the second embodiment is a program recommendation system that is communicably connected to a program viewing terminal 5 such as a program recording device, and a plurality of program viewing terminals 5 via a general-purpose network 7 such as the Internet. Consists of Server 6
- the configuration of the program recommendation system of the second embodiment is the same as the configuration of the program recommendation system shown in FIG. Originally, a configuration in which one program viewing terminal 5 is connected to the program recommendation server 6 will be described in order to simplify the explanation of the power to which a plurality of program viewing terminals 5 are connected.
- the program viewing terminal 5 receives a program from a broadcast wave or an IP network, views the program, records the program, plays the recorded program, deletes the recorded program, Make a recording reservation.
- the program viewing terminal 5 includes an antenna 701, a tuner 702, a TS decoder 703, CAS704, a signal number canceling unit 705, a demultiplexer 706, an MPEG Video decoder 707, an MPEG Audio decoder 708, an HDMI terminal 709, a recording encryption unit 713, a recording Control unit 714, recording medium 715, playback control unit 716, recording decoding unit 717, PSI extraction unit 718, EPG storage unit 719, key 720, infrared communication unit 721, marking management unit 722, marking storage unit 724, user identification unit 725, user information DB 726, marking transmission unit 727, recommended reception unit 728, recommended program information extraction unit 729 , A program record determination unit 730, a reservation management unit 731, a reservation memory 732, and a specific scene extraction
- the program recommendation server 6 includes a marking receiving unit 750, a user management unit 751, a user attribute DB 752, a marking registration unit 754, a marking DB 755, a specialization generation unit 70, a recommended scene creation unit 759, and a recommendation transmission unit. Consists of 760.
- the specialization generating unit 70 of the program recommendation server 6 includes a community marking statistical unit 756, an all-user marking statistical unit 757, and a community specializing unit 758.
- the tuner 702 receives broadcast waves of a predetermined channel from broadcast waves such as terrestrial digital broadcast and BS digital broadcast, or IP broadcast transmitted by multicast, and sends the broadcast waves to the TS decoder 703. Output.
- the TS decoder 703 demodulates the received broadcast wave in the general transport 'stream format and outputs it to the descrambling unit 705.
- CAS 704 is a general Conditional Access System, and reads out the encryption key information from the IC card in which the user registration information and the program encryption key are recorded, and outputs them to the descrambling unit 705.
- the descrambling unit 705 decrypts the encrypted broadcast signal demodulated by the TS decoder 703 based on the encryption key from the CAS 704, decrypts the encrypted program information and decrypts it. Output to multiplexer 706.
- the demultiplexer 706 separates information multiplexed as program information. At least video information, audio information, and program management information are multiplexed in the program information. After separating these, the video information is extracted to the MPEG Video decoder 707, the audio information is extracted to the MPEG Audio decoder 708, and the program management information is extracted to PSI. Output to part 718.
- the MPEG Video decoder 707 demodulates the video information from the demultiplexer 706, converts it into video data, and outputs it to the HDMI terminal 709.
- the MPEG Audio decoder 708 demodulates the audio information from the demultiplexer 706, converts it into audio data, and outputs it to the HDMI terminal 709.
- the HDMI terminal 709 is connected to an external HDMI terminal 710 via an HDMI cable.
- the HDMI terminal 710 is connected to a monitor 711 and a speaker 7 Connected with 12.
- HDMI terminals 709 and 710 are general video information and audio information interfaces.
- the program viewing terminal 5, the monitor 711, and the speaker 712 mutually authenticate each other and share an encryption key.
- the video data and audio data are encrypted with the shared encryption key and output to the monitor 711 and the speaker 712.
- the monitor 711 is a general program video information display device, which decrypts the encrypted video data received at the HDMI terminal 710 and outputs the video data of the program.
- the speaker 712 is a general program audio information output device that decrypts encrypted video data received at the HDMI terminal 710 and outputs program audio data.
- the recording encryption unit 713 performs a signal processing for recording the received program information on the recording medium.
- the recording encryption unit 713 encrypts the program information input from the demultiplexer 706 and outputs the encrypted program information to the recording control unit 714. Therefore, the recording encryption unit 713 performs encryption processing specific to the recording medium for recording the program information.
- DVD-RAM uses CP RM (Content Protection for Rewritable Media) and Blu-ray ice cream HD-DVD uses AACS (Advanced Access Content System).
- the recording control unit 714 controls the recording of the program information encrypted by the recording encryption unit 713 on the recording medium 715.
- the recording control unit 714 records program information on the recording medium by performing modulation processing and error correction encoding processing specific to the recording medium.
- Recording medium 715 is a computer-readable recording medium for recording program information.
- a magnetic recording disk such as a hard disk, an optical disk such as a DVD or a Blu-ray disk, or a semiconductor memory such as an SD card.
- magneto-optical recording media such as MO / MD are used.
- the reproduction control unit 716 reads program information recorded on the recording medium 715.
- the recording control unit 714, the recording medium 715, and the reproduction control unit 716 are generally configured as a drive system.
- the recording decryption unit 717 performs decryption corresponding to the encryption of the recording encryption unit 713.
- the recording decryption unit 717 decrypts the program information encrypted and recorded on the recording medium 715 and outputs the decrypted program information to the TS decoder 703.
- the program information output to the TS decoder 703 is reproduced.
- the live processing is the same as the playback processing of the program received by the antenna 701 described above.
- PSI extraction section 718 extracts PSI (Program Specific Information) from the program management information extracted by demultiplexer 706.
- PSI mainly contains information such as PMT (Program Map Table), EIT (Event Information Table), and TOT (Time Offset Table).
- the PMT stores packet identification information such as images and sounds included in the program! Therefore, if packet identification information such as images and sounds can be obtained from the PMT, the program can be reproduced by extracting packets with the packet identification information.
- the EIT includes information related to the program such as the program name, broadcast date and time, and broadcast content.
- EPG Electronic Program Guide
- the EPG storage unit 719 creates and stores EPG information using the EIT in the PSI extracted by the PSI extraction unit 718.
- Key 720 accepts input from the user, and is composed of a remote control and a front panel.
- the key 720 functions as a marking unit performed by the user for the program to be viewed.
- the remote control is provided with a marking button, and the program is marked while the user presses the marking button.
- the key 720 receives input of time information in a program to be marked from the user.
- the infrared communication unit 721 is a light receiving unit for infrared signals from a general remote controller. While the marking button attached to the key 720 is being pressed, the infrared communication unit 721 generates a marking button pressing signal and outputs it to the marking management unit 722.
- the marking management unit 722 receives the marking button press signal from the infrared communication unit 721 and determines that the marking button is pressed by the marking button press signal, the marking management unit 722 Time information is received from the PSI extraction unit 718 and output to the marking storage unit 724.
- the PSI extracting unit 718 includes a time extracting unit 7181 and a PID extracting unit 7182 therein.
- the time extraction unit 7181 extracts the playback time in the program by checking the time offset table (TOT) in the PSI.
- the PID extraction unit 7182 extracts the PID that is the packet identification information of the packet of the video content that is currently being played back, and identifies the packet that is currently being played back.
- the marking storage unit 724 stores V, time information marked in the program extracted by the marking management unit 722, or packet identification information group of the program being marked.
- the marking storage unit 724 may store time information in association with the time in the program at the start of marking and the time in the program at the end of marking, or the packet of the program packet at the start of marking. Even if the identification information and the packet identification information of the packet of the program at the end of marking are associated with each other and stored as a packet identification information group, it is possible to register the packet.
- the packet identification information of all packets being marked may be stored as one group.
- These time information and packet information are stored in the marking storage unit 724 in association with the program identification information of the program being reproduced.
- the program identification information is included in the PSI extracted by the PSI extraction unit 718.
- the marking position time information and packet identification information corresponding to the program identification information are stored in the marking storage unit 724 as marking information.
- the user identification unit 725 identifies a user who uses the program viewing terminal 5.
- the user identification unit 725 identifies a user who uses the program viewing terminal 5 based on user information input using the key 720.
- the user information DB 726 is a database in which user information of all users who use the program viewing terminal 5 is registered.
- the user identification unit 725 identifies a user based on user information input using the key 720 and user information stored in the user information DB 726 in advance. Further, when the user identification unit 725 succeeds in user authentication, the user identification unit 725 outputs user information of the user currently in use to the marking transmission unit 727.
- the marking transmission unit 727 is a general network interface, and after establishing a network with the program recommendation server 6 connected via the general-purpose network 7, the marking information stored in the marking storage unit 724 And the user information input from the user identification unit 725 are transmitted to the program recommendation server 6 together.
- marking storage unit 724 associates and stores the user who has been marked in conjunction with the user identification unit 725, and the marking transmission unit 727 transmits the marking information that has been transmitted to the program recommendation server 6 and the untransmitted marking information. Marking information is managed so that it can be identified.
- the recommendation receiving unit 728 recommends the program recommendation server 6 via the general-purpose network 7. Receive recommendation information about the program.
- the recommended program information extraction unit 729 extracts recommended program information from the recommendation information received by the recommendation receiving unit 728.
- the recommended program information is information that can uniquely identify a recommended program.
- the recommended program information includes program identification information for identifying a program, broadcast date / time of a program, channel information, and broadcast station identification information for identifying a broadcast station. Includes information.
- the recommended program information includes time information or packet identification information that can uniquely specify a recommended scene in the program.
- the extracted recommended program information is output to the program record determination unit 730.
- the program record determination unit 730 determines whether or not the program indicated by the program identification information is recorded in the recording medium 715 based on the program identification information indicated by the received recommended program information! Check. Further, when the program record determination unit 730 determines that the program indicated by the recommended program information is recorded, the recommended scene represented by the time information or the packet identification information in the program indicated by the recommended program information is recorded. Determine if! /. When determining that the recommended scene is recorded, the program recording determination unit 730 outputs the recommended program information to the specific scene extraction unit 733. Further, the program recording determination unit 730 determines that the program indicated by the recommended program information is recorded! /, NA! /, Or the recommended scene of the program is recorded! /, NA! /.
- the program record determination unit 730 In the case of refusal, it is determined based on the program identification information of the program and the EPG information stored in the EPG storage unit 719 whether the program is the current program or later. When it is determined that the program record determination unit 730 is a program scheduled to be broadcast after the present time, the program record determination unit 730 outputs the program identification information to the reservation management unit 731.
- the reservation management unit 731 performs recording reservation processing for the program indicated by the program identification information based on the program identification information from the program recording determination unit 730.
- the reservation memory 732 stores the program identification information of the program reserved by the reservation management unit 731 in association with the user information input from the user identification unit 725.
- the specific scene extraction unit 733 extracts the program indicated by the program identification information from the recording medium 715, and starts the recommended scene indicated by the recommended program information. Play the program from the time to the end time.
- the marking receiving unit 750 is a multi-program connected to the general-purpose network 7. From the marking transmission unit 727 of the viewing terminal 5, the marking information of the program marked by each user and the user information of the user who performed the marking are received.
- the marking information includes program identification information that can uniquely identify the marked program, or program broadcast date and time of the reserved program, channel information, and broadcast station identification information.
- the marking information includes time information in the marked program! /, And includes packet identification information.
- the received marking information and user information are output to the user management unit 751.
- User attribute DB752 contains registered user information and, if necessary, user name, gender, address / phone number, manufacturer / model of program viewing terminal, resolution of connected monitor, and network of program viewing terminal.
- User attribute information such as IDs is managed as a user registration list. The user performs user registration processing for program recommendation on the program recommendation server 6 in advance. In the user registration process, after user registration information such as user name setting and address / phone number notification is notified, the user information is generated and notified to the user on the program recommendation server 6 side. Store / manage user attribute information notified from the user in association with user information generated by DB752.
- the user management unit 751 determines whether or not the user specified by the received user information is a registered user of the program recommendation server 6.
- the user management unit 751 searches the user registration list stored in the user attribute DB 752 for user information that matches the received user information. When it is determined that the received user information is registered in the user attribute DB 752, it is determined that the user information is a registered user of the program recommendation server 6, and the processing is continued.
- the marking registration unit 754 is received by the marking reception unit 750 when the user management unit 751 determines that the user is a registered user of the program recommendation server 6. Marking information for each user is associated with the received user information.
- the community marking statistical unit 756 groups all users having marking information similar to the marking information as community groups.
- the community marking statistics unit 756 aggregates marking information other than the target program of the marking information of all users belonging to the community group, normalizes the counting result by the total number of users belonging to the community group, and Calculate the marking rate for each program.
- the community marking statistical unit 756 outputs the normalized marking rate to the community specializing unit 758.
- the similar marking information described above refers to the marking position in the program indicated by the marking information received by the marking receiving unit 750 and the marking in the program indicated by the marking information stored in the marking DB 755.
- the packet identification information in the program indicated by the received marking information and the packet identification information in the program indicated by the marking information stored in the marking DB 755 are overlapped. is there.
- the overlap between the marking position in the program indicated by the received marking information and the marking position in the program indicated by the marking information stored in the marking DB 755 is 50% of the total marking time of the received marking information. % Or more is preferable.
- the overlap between the packet identification information group in the program indicated by the received marking information and the packet identification information group in the program indicated by the marking information stored in the marking DB 755 is the total number of marking packets of the received marking information. Against 50% or more! / ,.
- the All User Marking Statistics Unit 757 has a user attribute DB752. All registered users, that is, all registered users in the program recommendation server 6 are grouped as all user groups including community groups.
- the all-user marking statistics unit 757 performs marking after grouping all users.
- the marking information of all users is statistically processed to create a marking frequency for each program. This marking frequency is normalized by the total time of all markings of registered users or the total number of marking packets.
- the created marking frequency is output to the community specialization unit 758 and the recommended scene creation unit 759.
- the community specialization unit 758 subtracts the marking frequency calculated by the all-user marking statistical unit 757 from the marking frequency calculated by the community marking statistical unit 756, thereby grouping the communities grouped by the community marking statistical unit 756.
- a marking frequency specialized for the group is created and output to the recommended scene creation unit 759.
- the marking frequency for each program is subtracted for each program.
- the recommended scene creation unit 759 Based on the marking frequency generated by the community specialization unit 758, the recommended scene creation unit 759 extracts a program having a marking frequency higher than a preset threshold as a recommended program. Also, the recommended scene creation unit 759 extracts, as recommended scene information, the start time and end time of the scene where the marking frequency exceeds the threshold value in each program, or packet identification information of the start packet and packet identification information of the end packet. .
- the recommended scene creation unit 759 identifies program identification information for identifying a recommended program, or broadcast date and time of the recommended program, channel information on which the recommended program is broadcast, and a broadcasting station on which the recommended program is broadcast.
- the recommended program information including broadcast station identification information and recommended scene information is created. Accordingly, the recommended scene creation unit 759 creates a recommended scene specialized for the community group.
- the recommended scene creation unit 759 extracts a program having a marking frequency higher than a preset threshold! / As a popular program based on the marking frequency generated by the all-user marking statistics unit 757. To do. Also, the recommended scene creation unit 759 extracts the start time and end time of a scene whose marking frequency exceeds a threshold in each program, or packet identification information of the start packet and packet identification information of the end packet as popular scene information. To do. The recommended scene creation unit 759 broadcasts program identification information for identifying popular programs, broadcast date and time of popular programs, channel information for broadcasting popular programs, and popular programs. Popular program information including broadcast station identification information for identifying a broadcast station to be broadcast and popular scene information is created.
- recommended program information including the start time and end time of the recommended scene specialized for the community group and including the program identification information of the program including the recommended scene is generated, and the popular scene starts.
- the popular program information including the time and end time and including the program identification information of the program including the popular scene is generated and output to the recommended transmission unit 760.
- the recommended transmission unit 760 transmits the recommended program information created by the recommended scene creation unit 759, including recommended program information including recommended scene information, and popular program information including popular scene information, to the program viewing terminal 5. .
- the user A marks a part of the program A by the program viewing terminal 5 owned by the user A.
- the program viewing terminal 5 provides the program recommendation server 6 with user information for identifying the user A, program identification information for identifying the program A, and marking information for identifying the marked position in the program. Send.
- the marking receiving unit 750 of the program recommendation server 6 is a program viewing terminal owned by the user A.
- the user management unit 751 determines whether or not the received user information is stored in the user attribute DB 752, and if it is stored, the processing is continued assuming that the user A is a registered user of the program recommendation server 6.
- the marking registration unit 754 associates with the user information for identifying the user A, the program identification information for identifying the program A, the marking information for identifying the marked position in the program, Register the marking in DB755.
- the Community Marking Statistics Department 756 extracts all users with marking information similar to the marking information of Program A from the marking DB 755 to form a community group.
- the community marking statistics unit 756 statistically processes marking information of programs other than program A in the community group.
- the community marking statistics unit 756 calculates the marking frequency of programs other than program A in the community group. , Output to Community Specialization Department 758.
- the all-user marking statistical unit 757 extracts all users registered in the user attribute DB 752 of the program recommendation server 6 from the marking DB 755 and includes all user groups including the community group formed by the community marking statistical unit 756. Form.
- the all-user marking statistics unit 757 extracts marking information associated with user information in all user groups from the marking DB 755, and statistically processes the marking information of programs other than program A. Then, the all-user marking statistical unit 757 calculates the marking frequency of programs other than the program A in all user groups, and outputs it to the community specializing unit 758 and the recommended scene creating unit 759.
- the community specializing unit 758 subtracts the marking frequency of programs other than the program A created by the all-user marking statistical unit 757 from the marking frequency of programs other than the program A created by the community marking statistical unit 756. . As a result, the community specializing unit 758 creates a marking frequency of the program specialized for the community group to which the user who marked a predetermined position in the program A belongs, and outputs it to the recommended scene creating unit 759.
- the recommended scene creation unit 759 has a marking frequency higher than the threshold value based on the marking frequency specialized for the community group created by the community specialization unit 758!
- the recommended program information is created by extracting the recommended scene information indicating the higher marking frequency! / And the scene start time and end time.
- the recommended scene creation unit 759 also includes program identification information of a program with a marking frequency higher than the threshold based on the generalized marking frequency of all registered users created by the all-user marking statistics unit 757, and the threshold value.
- the popular program information is created by extracting the popular scene information indicating the start time and end time of the scene with higher marking frequency! /.
- the recommendation transmission unit 760 transmits the recommended program information and popular program information created by the recommended scene creation unit 759 to the program viewing terminal 5 of the user A.
- FIGS. 13 to 15 are flow charts showing the processing flow of the program recommendation system according to the second embodiment, and FIG. 13 is the marking information in the program viewing terminal according to the second embodiment.
- FIG. 14 is a flowchart showing a flow of processing in the program recommendation server according to the second embodiment, and FIG. 15 is a program according to the second embodiment.
- 6 is a flowchart showing a flow of processing after receiving recommendation information in the viewing terminal.
- the user identification unit 725 displays a user authentication screen for user authentication for the user on the monitor 711, and the key 720 accepts input of user information by the user, and the program viewing terminal 5 The login to is accepted (step S41).
- the user identification unit 725 determines whether or not the user authentication is successful, that is, whether or not the input user information matches the user information registered in the user information DB 726 in advance ( Step S42).
- the user identification unit 725 outputs an error message to the monitor 711 and ends the subsequent processing (step S43).
- step S42 if it is determined that the registered user of the program viewing terminal 5 has no error in the input user information (YES in step S42), the playback control unit 716 performs processing for the program recorded on the recording medium 715. It is determined whether or not a reproduction instruction operation has been performed (step S44). Here, when it is determined that there is no reproduction instruction operation (NO in step S44), reproduction control unit 716 waits until there is a reproduction instruction operation from the user.
- playback control unit 716 plays back the program selected by the user from recording medium 715 and outputs it to monitor 711. Playback processing is executed (step S45).
- the marking management unit 722 determines whether or not the marking operation by the user has been started (step S46).
- the infrared communication unit 721 senses an infrared signal output when the marking button of the remote controller is pressed, it generates a marking button pressing signal and outputs it to the marking management unit 722.
- the marking management unit 722 determines that the marking operation has been started when the marking button pressing signal is input.
- the time extraction unit 7181 extracts the playback time in the program from the PSI multiplexed in the program broadcast, and performs the marking operation.
- step S47 Is temporarily stored in the marking storage unit 724 as the marking start time when the operation is started (step S47). On the other hand, if it is determined that the marking operation has not started (NO in step S46), the process proceeds to a determination process for determining whether or not the program being played has ended (step S50).
- the marking management unit 722 determines whether or not the marking operation by the user is completed (step S48).
- the infrared communication unit 721 ends the generation of the marking button pressing signal.
- the marking management unit 722 determines that the marking operation is finished when the marking button pressing signal is not input.
- step S48 If it is determined that the marking operation has been completed (YES in step S48), the time extraction unit 7181 reads the program from the PSI multiplexed in the program in the same manner as when the marking start time is extracted. The reproduction time is extracted and temporarily stored in the marking storage unit 724 as the marking end time when the marking operation is completed (step S49). On the other hand, when it is determined that the marking operation has not been completed (NO in step S48), the marking management unit 722 waits until the marking operation is completed.
- the marking transmission unit 727 determines whether or not the program being reproduced has ended (step S 50). If it is determined that the program is finished (YES in step S 50), the marking transmission unit 727 reads the marking start time, marking end time, and playback performed in the marking storage unit 724. The program identification information of the program and the user information of the user identified by the user identification unit 725 are transmitted to the program recommendation server 6 as marking information (step S51). On the other hand, if it is determined that the program being played has not ended (NO in step S50), the playback process in step S45 is continued.
- the marking receiving unit 750 of the program recommendation server 6 receives the marking information transmitted by the marking transmitting unit 727 of the program viewing terminal 5 (step S61).
- the user management unit 751 determines whether or not the user authentication has succeeded, that is, marking. It is determined whether or not the user information included in the marking information received by the receiving unit 750 matches the user information registered in advance in the user attribute DB 752 of the program recommendation server 6 (step S62).
- the user management unit 751 creates an error message to that effect and causes the program viewing terminal 5 to receive an error. A message is transmitted, and the subsequent processing is terminated (step S63).
- the program viewing terminal 5 that has received the error message outputs to the monitor 711 an error notification indicating that the user authentication has failed.
- the marking registration unit 754 uses the marking information received by the marking reception unit 750 as the user's user.
- the marking is associated with the information and registered in the marking DB755 (step S64).
- the community marking statistics unit 756 extracts all the users with marking information similar to the marking information received by the marking receiving unit 750, and groups them as community groups.
- the similar marking information is the same program identification information as the program identification information indicated by the marking information received by the marking receiving unit 750, and overlaps with the marked time in the program and the packet identification information of the marked program. This is marking information.
- marking information that overlaps 50% or more of the marked time and marked packet identification information for each user in the same program is determined as similar marking information.
- the community marking statistics unit 756 includes the marking received by the marking receiving unit 750 among the marking information registered for all users in the community group composed of users having similar marking information. Marking information other than the program indicated by the information is aggregated for each program (step S66).
- the community marking statistics unit 756 counts, if there is a user marking for each unit time in each program, and calculates the frequency distribution on the time axis for each program.
- the community marking statistic unit 756, if there is a user who performed marking for each packet in each program calculates the frequency distribution in units of packets for each program.
- the community marking statistics unit 756 normalizes the aggregated results by the total marking time or the total number of marking packets of all users grouped in the community group (step S67).
- the community marking statistics unit 756 outputs the normalized result to the community specialization unit 758.
- the all-user marking statistics unit 757 extracts users registered in the user attribute DB 752, that is, all users registered in the program recommendation server 6, and has similar marking information. Grouping is performed as all user groups including community groups composed of users (step S68).
- the all-user marking statistics unit 757 extracts the marking information of all users from the marking DB 755, and totals the marking information of programs other than the program indicated by the marking information received by the marking receiving unit 750. (Step S69).
- the total user marking statistics unit 757 counts, if there is a user marking every unit time in each program, and calculates the frequency distribution on the time axis for each program.
- the total user marking statistics unit 757 counts, if there is a user who marked each packet in each program, and calculates the frequency distribution in units of packets for each program.
- the total user marking statistical unit 757 normalizes the aggregated result with the total marking time of all registered users! /, Based on the total number of marking packets (step S 7 0). ).
- the normalized result is output to the community specializing unit 758 and the recommended scene creating unit 759.
- the community specializing unit 758 performs normalized normalization within all user groups based on the normalized marking aggregation result (marking frequency) within the community group composed of users having similar marking information. Subtract the marking result (marking frequency). Then, the recommended scene creation unit 759 determines the threshold value of the subtraction result, and indicates the marking frequency higher than the threshold! /, The program identification information of the program, and the recommended start time and end time of the scene with the marking frequency higher than the threshold value.
- the recommended program information is created by extracting the scene information (step S71).
- the start time is the time position where the marking frequency is above the threshold
- the end time is the time position where the marking frequency is below the threshold.
- the start time and end time of all the extracted recommended scenes correspond to the corresponding program identification information.
- it is output to the recommended transmitter 760 as recommended program information.
- the recommended scene creation unit 759 determines the threshold value of the normalized marking count result in all user groups created by the all user marking statistics unit 757, and the marking frequency is higher than the threshold value! /
- the popular program information is created by extracting the program identification information of the program and the popular scene information indicating the start time and end time of the scene having the marking frequency higher than the threshold (step S72).
- the extracted start time and end time of all popular scenes are output to the recommended transmission unit 760 as popular program information in association with corresponding program identification information.
- the recommendation transmitting unit 760 transmits the recommended program information and popular program information created by the recommended scene creating unit 759 as recommended information to the program viewing terminal 5 (step S73).
- the recommendation receiving unit 728 of the program viewing terminal 5 receives the recommended information including the recommended program information and the popular program information transmitted by the program recommendation server 6 (step S81). .
- the recommended program information extraction unit 729 determines whether or not the recommended information received by the recommendation receiving unit 728 and the recommended program information have been extracted (step S82).
- the recommended program information extraction unit 729 prompts the user to select between the recommended program information and the popular program information! /.
- the recommended program information extraction unit 729 extracts the recommended program information
- the popular program information is selected, the recommended program information extraction unit 729 extracts the popular program information. To do. It should be noted that setting for extracting both recommended program information and popular program information does not matter. As a result, it is possible to provide the user with more options.
- the program record determination unit 730 uses the recommended scene information included in the recommended program information extracted by the recommended program information extraction unit 729. Based on the program identification information corresponding to the accompanying recommended scene, it is determined whether or not the program including the recommended scene has been recorded on the recording medium 715 (step S83). In step S83, it is preferable to determine whether or not the recommended scene itself is recorded on the recording medium 715 based on the start time and end time of the recommended scene as well as the program recording. Here, the program including the recommended scene, or the recommended scene If it is determined that no object is recorded on the recording medium 715 (NO in step S83), the process proceeds to step S88.
- the specific scene extraction unit 733 displays the recommended program on the monitor 711. Display (step S84). At this time, the specific scene extraction unit 733 may display the start time and end time of the recommended scene, the time length of the recommended scene, and the captured image or video in the recommended scene, in addition to the program name.
- the program recommendation server 6 according to the frequency distribution of each recommended scene created by the recommended scene creation unit 759, based on the value obtained by integrating the frequency distribution for each recommended scene or the peak frequency of the frequency distribution.
- a priority may be assigned to each recommended scene and transmitted to the program viewing terminal 5.
- the specific scene extracting unit 733 may display the recommended scene with the size of the text in proportion to the priority or the size of the gap. This allows the user to determine the priority of the program scene at a glance.
- the specific scene extraction unit 733 determines whether or not a user has issued a recommended scene selection and playback instruction (step S85). If it is determined that the recommended scene playback instruction has not been issued (NO in step S85), the process proceeds to step S88. On the other hand, if it is determined that the recommended scene playback instruction has been issued (YES in step S85), the specific scene extraction unit 733 determines the program identification information and recommended scene of the program selected by the user from the recommended program information. And the start time of.
- the playback control unit 716 reads program information corresponding to the program identification information extracted by the specific scene extraction unit 733 from the recording medium 715, and recommends the recommended scene from the start time extracted by the specific scene extraction unit 733. Is played (step S86).
- the specific scene extraction unit 733 determines whether or not the recommended scene reproduction processing has ended (step S87). If it is determined that the recommended scene playback process has been completed! /, NA! / (NO in step S87), the specific scene extraction unit 733 waits until the recommended scene playback process ends. [0227] On the other hand, if it is determined that the recommended scene playback process has been completed (YES in step S87), the specific scene extraction unit 733 displays all recommended scenes included in the recommended program information transmitted by the program recommendation server 6. It is determined whether or not the information reproduction process is completed (step S 88). When it is determined that all the recommended scene information has been reproduced! /, NA!
- step S88 the process returns to step S83, and the remaining recommended scene information is processed. .
- step S88 the process returns to step S83, and the remaining recommended scene information is processed.
- program record determination unit 730 extracts recommended program information. Whether or not the program including the popular scene has been recorded on the recording medium 715 based on the program identification information corresponding to the popular scene associated with the popular scene information included in the popular program information extracted by the section 729. (Step S89). In step S89, it is preferable to determine whether or not the popular scene itself is recorded on the recording medium 715 based on the start time and end time of the popular scene as well as the program recording. If it is determined that the program including the popular scene or the popular scene itself has not been recorded in the recording medium 715 (NO in step S89), the process proceeds to step S94.
- the specific scene extraction unit 733 displays the popular program on the monitor 711. Display (step S90). At this time, the specific scene extraction unit 733 may display the start time and end time of the popular scene, the time length of the popular scene, and a captured image or video in the popular scene, simultaneously with the program name.
- the frequency distribution for each popular scene is integrated or based on the peak frequency of the frequency distribution. Priorities may be assigned to each popular scene and transmitted to the program viewing terminal 5.
- the specific scene extraction unit 733 may display the popular scene with the size of text proportional to the priority or the size of the captcha. This allows the user to determine the priority of the program scene at a glance.
- the specific scene extraction unit 733 determines whether or not the user has issued a selection and playback instruction for the popular scene (step S91).
- step S91 If it is determined that the popular scene playback instruction has not been issued (NO in step S91), the process proceeds to step S94.
- the specific scene extraction unit 733 determines the program identification information and popular scene of the program selected by the user from the popular program information. And the start time of. Then, the playback control unit 716 reads program information corresponding to the program identification information extracted by the specific scene extraction unit 733 from the recording medium 715, and starts the popular scene from the start time extracted by the specific scene extraction unit 733. Is played (step S 92).
- the specific scene extraction unit 733 determines whether or not the popular scene reproduction processing has ended (step S93). If it is determined that the popular scene playback process has ended! /, NA! / (NO in step S93), the specific scene extraction unit 733 waits until the popular scene playback process ends.
- step S93 when it is determined that the popular scene playback process has ended (YES in step S93), the specific scene extraction unit 733 displays all popular scenes included in the popular program information transmitted by the program recommendation server 6. It is determined whether or not the information reproduction process has been completed (step S 94). If it is determined that all popular scene information playback processing has been completed! /, NA! / (NO in step S94), the process returns to step S89, and the remaining popular scene information is processed. . On the other hand, if it is determined that all popular scene information playback processing has ended (YES in step S94), all processing ends.
- step S83 in addition to determining whether a program including a recommended scene has been recorded! /, The program record determination unit 730 has a program including a recommended scene! / If it is determined that the user has played back, the recommended scene may not be played back, and the next recommended scene may be played back. To this end, a replay flag is assigned to the portion reproduced for the program recorded on the recording medium 715 together with the user information of the replayed user in association with the time axis. This allows each user in the program to If the recommended scene has already been reproduced! /, It is not necessary to display the recommended scene in step S84.
- the user directly marks the program preference scene, the preference scene for each user is managed by the program recommendation server 6, and similar marking information is obtained. Grouping users, generating a marking frequency within the group, and recommending other scenes relating to other preferences to the user based on the marking frequency. Therefore, the user can efficiently view only the minimum necessary scenes than when the program itself is introduced. Also, even if there are programs that are marked by almost all users, such as popular programs, when the recommended program information in the group is generated, the viewing tendency of the entire user is deleted, and It is possible to recommend scenes in the program specialized for The.
- FIG. 16 is a diagram illustrating an example of a data structure of marking information transmitted / received between the marking transmission unit 727 and the marking reception unit 750.
- the marking information shown in FIG. 16 is the data structure when the program specified by the user power S specified by the user information “A” and the program ID “T” is marked for 7 minutes from the start of the program.
- the program viewing terminal 5 transmits to the program recommendation server 6 marking information including user information “A”, program ID “1”, and time information indicating that the program has been marked for 7 minutes from the start.
- FIG. 17 is a diagram showing an example of a data structure of marking information for each registered user of the program recommendation server 6 stored in the marking DB 755. Since the marking information including the user ID “A” and the program ID “T” has been sent now, it is confirmed by the user management unit 751 whether or not the user ID “A” is registered in the user attribute D B752. . When the user ID “A” is registered, the received marking information shown in FIG. 16 is registered in the marking DB 755. As a result, as shown in FIG. 16, marking information 1001 relating to the user ID “A” is added to the marking DB 755.
- the marking information 1002 includes a user ID “B”, a program ID 'T', and time information indicating that the program has been marked for 2 to 8 minutes.
- 03 includes user ID “C”, program ID “1”, and time information indicating that the program has been marked for 5 to 12 minutes.
- Marking information 1004 includes user ID “C” and program ID. Includes 'T' and time information indicating that the program was marked twice from 1 to 8 minutes and from 16 to 25 minutes
- the community marking statistical unit 756 extracts users who have the same thread and marking as the received marking information, and determines the similarity of the marking information for grouping.
- FIG. 18 is a timing chart showing the marking information shown in FIG. 17 on the time axis.
- the marking time 1101 in the program “1” of the user “A” is 7 minutes from the start position of the program to 7 minutes later.
- Marking The marking time 1102 for the user “B” program “1” stored in the DB 755 is 7 minutes from 2 minutes to 9 minutes after the program start position. Therefore, the marking time 1101 for user “A” and the marking time 1102 for user “B” overlap each other by 50% or more, and user “A” and user “B” are grouped into the same group.
- the marking time 1101 in the program “1” of the user “A” is 7 minutes from the start position of the program to 7 minutes later.
- Marking The marking time 1102 for the user “B” program “1” stored in the DB 755 is 7 minutes from 2 minutes to 9 minutes after the program start position. Therefore, the marking time 1101 for user “A” and the marking time 1102 for user “B” overlap each other by 50% or more,
- the marking time 1103 in the program “1" of the user “C” stored in the marking DB 755 is 7 minutes from 5 minutes to 12 minutes after the program start position. Therefore, the marking time 1101 for user “A” and the marking time 1103 for user “C” do not overlap each other by 50% or more, and user “C” is not grouped into user “A” group. Yes.
- the marking time 1104 in the program “1" of the user “D” stored in the marking DB755 is 7 minutes from 1 minute to 8 minutes after the program start position, and the marking time 1105 is , 7 minutes from 16 minutes to 25 minutes after the start of the program. Therefore, the marking time 1101 for user “A” and the marking time 1104 for user “D” have an overlap of 50% or more, and user “D” is grouped into the group of user “A”.
- FIG. 19 is a diagram illustrating an example of a data structure of the marking DB 755 in which marking information including program IDs other than the program ID 'T' is stored.
- the data structure of the marking DB 755 shown in Fig. 19 has the marking information similar to the marking information of the user "A” as described above, and the marking information of the user "B” and the user “D” grouped together.
- marking information of the user “Z” is stored, and marking information including a program ID other than the program ID “1” is stored.
- program ID "2" and program ID "4" are associated with user ID "A”.
- the marking information 1201 includes a user ID “A”, a program ID “2”, and time information indicating that the program has been marked from 0 to 40 minutes.
- the marking information 1202 includes a user ID “A”, It includes program ID “4” and time information indicating that the program has been marked for 30-70 minutes and 90-; 110 minutes.
- the marking information 1203 includes user ID “B”, program ID “2”, and time information indicating that the program has been marked for 10 to 50 minutes.
- Marking information 1204 includes user ID “B”. , Including program ID “3” and time information indicating that the program has been marked for 10 to 50 minutes.
- Marking information 1205 includes user ID “B”, program ID “4”, and program number 30. And time information indicating marking up to 70 minutes.
- Marking information 1206 includes user ID “Z”, program ID “3”, and time information indicating that the program has been marked for 10 to 50 minutes, and marking information 1207 includes user ID “Z”.
- Marking information 1208 includes user ID “D”, program ID “2”, and time information indicating that the program has been marked for 20 to 40 minutes.
- Marking information 1209 includes user ID “D”, program ID It includes ID “3” and time information indicating that the program has been marked for 20 to 30 minutes.
- the marking information 1210 includes user ID “D”, program ID “4”, program 40 to 50 minutes and Including time information indicating marking from 100 to 120 minutes. Mu
- FIG. 20 is a conceptual diagram for explaining a method of counting marking information in the drape in the community marking statistical unit 756 of the program recommendation server 6 of the present program recommendation system.
- the first 0 to 60 minutes is the program ID "2" playback time
- 0 to 60 minutes is the program ID "3" playback time
- 0 to 120 minutes is the program ID. It represents the playback time of “4”.
- Marking times 1301, 1302, and 1303 shown in FIG. 20 indicate the marking times of the user “A”.
- the marking time 1301 is 40 minutes from the start of the program “2” shown by the marking information 1201 in FIG. 19 to 40 minutes later.
- the marking time 1302 is 40 minutes from 30 minutes to 70 minutes after the program start position of the program “4” indicated by the marking information 1202 in FIG.
- the marking time 1303 is 20 minutes from 90 minutes to 110 minutes after the program start position of the program “4” indicated by the marking information 1202 in FIG.
- Marking times 1304, 1305, and 1306 shown in FIG. 20 indicate the marking times of the user “B”.
- the marking time 1304 is 40 minutes from 10 minutes to 50 minutes after the program start position of the program “2” indicated by the marking information 1203 in FIG.
- the marking time 1305 is 40 minutes until 50 minutes after 10 minutes after the program start position of the program “3” indicated by the marking information 1204 in FIG.
- the marking time 1306 is 40 minutes from 30 minutes to 70 minutes after the program start position of the program “4” indicated by the marking information 1205 in FIG.
- Marking times 1307 and 1308 shown in FIG. 20 indicate the marking times of the user “Z”.
- the marking time 1307 is 40 minutes from 10 minutes to 50 minutes after the program start position of the program “3” indicated by the marking information 1206 in FIG.
- the marking time 1308 is 40 minutes from 20 minutes to 60 minutes after the program start position of the program “4” indicated by the marking information 1207 in FIG.
- Marking time 1309 is indicated by marking information 1208 in Fig. 19. It is 20 minutes from 20 minutes to 40 minutes after the program start position of program “2.”
- Marking time 1310 is 20 of the program start position of program “3” shown by marking information 1209 in FIG. 10 minutes from 30 minutes to 30 minutes.
- the marking time 1311 is 10 minutes from 40 minutes to 50 minutes after the program start position of the program “4” indicated by the marking information 1210 in FIG.
- the marking time 1312 is 20 minutes from 100 minutes to 120 minutes after the program start position of the program “4” indicated by the marking information 1210 in FIG.
- the marking frequency distribution of FIG. 20 shows the statistical results of the marking information of the programs “2”, “3” and “4” of the community group.
- the statistical processing in this embodiment is totaled so as to form a frequency distribution by counting the number of markings for each user in unit time (10 minutes in this example).
- This frequency distribution is normalized by the total marking time of users in the community group, and then the normalized marking statistics for all user groups are subtracted to generate a marking frequency distribution specialized for the community group. Is done.
- the normalized statistical results for programs “2”, “3”, and “4” are calculated as a distribution that is close to 0 at any unit time. Subtracting the result is almost the same as the marking frequency distribution within the community group.
- the recommended scene is generated by the recommended scene creation unit 759 by performing threshold determination using a predetermined threshold 1316 based on the normalized marking frequency distribution in the community group.
- program scene 1313 for program "2" that exceeds threshold 1316 for 20 to 40 minutes, program scene 1314 for program "3" for 20 to 30 minutes, and program "4" for 30 to 50 minutes A program scene is created as a recommended scene. Then, recommended information including time information for specifying a recommended scene and program identification information for identifying a program including the recommended scene is transmitted to the program viewing terminal 5. Also, priorities are given in the order of program scene 1315, program scene 1313, and program scene 1314 having the highest integrated value of the frequency distribution of each program scene.
- the recommendation receiving unit 728 of the program viewing terminal 5 receives this recommendation information, and the program recording determination unit 730 determines whether each recommended scene has been recorded in the recording medium 715, and the recording medium 71 5 Those not recorded in are deleted from the recommendation information. Further, the recording medium 715 also deletes the recommended scene that has already been reproduced by the user and does not display the recommended scene. In this example, the recommended scene 1313 and the recommended scene 1315 have already been reproduced by the user “ ⁇ ”, and therefore the recommended scene 1313 and the recommended scene 1315 are recommended for the user “ ⁇ ”. Not recommended.
- FIG. 21 is a diagram illustrating an example of the data structure of the marking DB of the second application example of the second embodiment.
- a user having marking information similar to the marking information for the user program “1” received by the marking receiving unit 750 is grouped to display a program for all users in the group. Indicates marking information for programs other than “1”.
- the data structure shown in FIG. 21 is different from the data structure shown in FIG. 19 in that a flag 1400 indicating the type of user operation is provided.
- the data structure of DB755 has the flag 1400 as described above.
- the marking information transmitted from the program viewing terminal 5 to the program recommendation server 6 is given a flag indicating the type of operation for the program. For example, when marking 10 to 50 minutes of program “A”, program identification information indicating program “A”, marking time information or packet identification information indicating 10 to 50 minutes, and marking operation are performed. Marking information consisting of identification information is sent. In this case, the user operation is not limited to the marking operation. Identification information for identifying operation information such as reproduction, deletion, or fast-forwarding may be added.
- the data structure of FIG. 21 is the user “B” with marking information similar to the marking information representing the marking to the user “A” program “1” received by the marking receiver 750, This is a result of grouping “C” and “D” and extracting marking information other than program “1” from the marking DB 755 within the group.
- User ID "A” is associated with program ID "2" in addition to program ID 'T'. Marking information 1401 is fast-forwarded for 0 to 5 minutes, 15 to 35 minutes, and 50 to 80 minutes of program “2”, and 5 to; 15 minutes, 35 to 50 minutes, and 80 to 95 minutes are played back. Marking time and flag.
- “F” represents a fast forward operation
- “P” represents a reproduction operation
- “M” represents a marking operation
- “D” represents an erasing operation.
- User ID "B” is associated with program ID "2” in addition to program ID 'T'. Marking information 1402 is a mark indicating that the program “2” is operated for 30-35 minutes, 45-65 minutes and 70-95 minutes, 35-45 minutes are marked, and 65-70 minutes are fast-forwarded. Time and flag.
- program ID "2" is associated with user ID "C”.
- Marking information 1403 is the program "2" 0 ⁇ ; 10 minutes, 20 ⁇ 35 minutes, 40 ⁇ 85 minutes and 90 ⁇ ; 125 minutes are played back, 10 ⁇ 20 minutes, 35 ⁇ 40 minutes and 85 ⁇ 90 The minute has been marked and has V, the marking time and flag.
- User ID "D" is associated with program ID "2" in addition to program ID 'T'.
- Marking information 1404 is the program "2" 0-25 minutes, 30-65 minutes, 70-85 minutes and 90-; 125 minutes are played back, 25-30 minutes and 65-70 minutes are deleted, Has a marking time and flag that 85-95 minutes have been marked.
- FIG. 22 is a conceptual diagram for explaining a method of counting marking information grouped as shown in FIG.
- the stream shown in FIG. 22 virtually shows a stream on the time axis of program “2”.
- CM Common Message
- a marking time 1504 shown in FIG. 22 indicates a marking time in program “2” which is a program other than program “1” of user “A” shown in marking information 1401 in FIG.
- fast-forward operation is performed for 0 to 5 minutes, 15 to 35 minutes and 50 to 80 minutes, 5 to; playback operations are performed for 15 minutes, 35 to 50 minutes and 80 to 95 minutes. Yes.
- the marking time 1505 shown in FIG. 22 is the user shown in the marking information 1402 of FIG.
- Program "B” is a program other than "B" program "1"
- the marking time is shown. At the marking time 1505, 30 to 35 minutes, 45 to 65 minutes and 70 to 95 minutes are played back, 35 to 45 minutes are marked, and 65 to 70 minutes are fast forwarded.
- the marking time 1506 shown in FIG. 22 is a program “2” that is a program other than the program “1” of the user “C” shown in the marking information 1403 of FIG.
- the marking time is shown. Marking time 1506, 0-10 minutes, 20-35 minutes, 40-85 minutes and 90-; 125 minutes (regeneration operation drowning, 10-20 minutes, 35-40 minutes and 85-90 minutes A marking operation is being performed.
- the marking time 1507 shown in FIG. 22 is a program “2” that is a program other than the program “1” of the user “D” shown in the marking information 1404 of FIG.
- the marking time is shown. At marking time 1507, 0 to 25 minutes, 3 0 to 65 minutes, 70 to 85 minutes and 90 to 85; 125 minutes for playback operations, 25 to 30 minutes and 65 to 70 minutes for deletion operations, 85 Marking operation is performed for ⁇ 95 minutes.
- the marking frequency distribution of Fig. 22 shows the result of counting the marking information of the community group.
- the tabulation processing in this embodiment is tabulated by adding different points for each piece of operation information indicated by the marking information in the unit time of the program.
- points for each piece of operation information indicated by the marking information in the unit time of the program For example, "2" points for the erase flag "D", "1" for the fast-forward flag "F"
- the marking information of programs other than program "1" is also aggregated in the same manner for all user groups configured by all registered users of the program recommendation server 6 including community groups.
- the operation flag of all users is given! /, And the total marking time is normalized. Then, by subtracting the normalized aggregated results of all user groups from the normalized aggregated results of community groups, aggregated results specific to the community group are created.
- the normalized statistical results for all user groups are The unit time is almost “0”, and the subtraction result and the subtraction result are almost the same.
- a recommended scene is generated by performing a threshold determination on the subtracted count result using a predetermined threshold.
- program scene viewing recommendation information and program scene deletion recommendation information are created using two threshold values.
- the recommended scene creation unit 759 extracts the marking time information whose normalized and subtracted tabulation results exceed the upper first threshold value 15 08 to view the program scene viewing recommended scenes 1510, 15 11 and 1512. Generate recommendation information. Also, the recommended scene creation unit 759 extracts the marking time information for which the normalized and subtracted counting result is lower than the second threshold value 1509 on the lower side, and generates deletion recommendation information for the program scene deletion recommended scenes 1513 and 1514. To do.
- the viewing recommendation information and the deletion recommendation information created as described above are recommended information including the program identification information of each program and the start time and end time of each program scene, by the recommended transmission unit 760. Sent to the program viewing terminal 5.
- the transmitted recommendation information is received by the recommendation receiving unit 728 of the program viewing terminal 5, and based on the program identification information included in the recommendation information, whether or not the program is recorded on the recording medium 715 is determined. A determination is made. If it is determined that the program scene is recorded, the program recording determination unit 730 further determines whether or not the program scene of each program is recorded based on the start time and end time of each program scene. Also, the program recording determination unit 730 automatically deletes the program scene when it is determined that the program scene has been recorded and the program scene is a deletion recommended scene. Further, when it is determined that the program scene has been recorded and the program scene is a viewing recommendation scene, the program recording determination unit 730 extracts program information based on the start time of the program scene. Then, a thumbnail is created and displayed on the monitor 711 to prompt the user to view it.
- the program viewing terminal 5 when the program viewing terminal 5 receives the program scene deletion recommendation information and deletes the program scene, the program viewing terminal 5 deletes the time position in the program where the deletion has started and the program within the program where the deletion has ended. It is desirable to connect the time position.
- FIG. 23 is a conceptual diagram showing a schematic configuration of the program recommendation system of the third embodiment.
- one screen 1600 in the program is divided into a plurality of clusters 160;!
- Each of the clusters 1601 to 1604 is broadcasted selectably by a pointing device 1605 such as a laser pointer or a mouse.
- the clusters 1602 to 1604 are broadcast in a form including one object such as an actor or a product in the program.
- the coordinate information of each cluster is multiplexed with the program information.
- the coordinate information can be delivered to the program viewing terminal 8 using the network 10 from the cluster information providing server that provides the coordinate information of the cluster in synchronization with the broadcast of the program.
- the program viewing terminal 8 transmits the program identification information of the broadcast being received to a cluster information providing server (not shown), and receives the cluster information corresponding to the transmitted program identification information.
- the user uses the pointing device 1605 to select a favorite object from among the objects displayed on the screen.
- a laser pointer is used as a pointing device
- a photosensitive film is attached to the surface of the monitor screen connected to the program viewing terminal 8, and coordinate information irradiated with the laser is extracted.
- a marker such as an arrow is displayed on the screen, and the coordinate information of the place where the user clicked is extracted.
- Cluster information multiplexed in the program is recorded in the cluster ID storage unit 1606.
- the program viewing terminal 8 determines which cluster the coordinate information belongs to based on the coordinate information selected by the pointing device 1605, and obtains cluster identification information based on the multiplexed cluster information.
- the cluster identification information is acquired and transmitted to the program recommendation server 9 connected by the general-purpose network 10.
- the program recommendation server 9 extracts users having the same cluster identification information from the received cluster identification information from its own cluster DB and groups them.
- the program recommendation server 9 aggregates all the selected cluster identification information associated with the grouped users with respect to all users in the group, forms a ranking, and performs ranking. Based on the above, a recommended cluster is generated. Then, the program recommendation server 9 transmits the recommended cluster to the program viewing terminal 8.
- the cluster search unit 1608 of the program viewing terminal 8 searches the recording medium (recording means) 1607 for a program including the cluster, and selects it as a recommended cluster screen 1609. To prompt the user to select.
- the program is played from the specified playback time position of the program that includes the recommended cluster.
- the recommended cluster may be displayed in a display form based on the recommended priority, for example, by changing the size of the thumbnail or the color. In FIG. 23, the thumbnail size is changed according to the number of users who have selected recommended clusters.
- FIG. 25 is a block diagram showing a configuration of a program viewing terminal according to the third embodiment
- FIG. 25 is a block diagram showing a configuration of a program recommendation server according to the third embodiment
- the program recommendation system according to the third embodiment includes a program recommendation server that is communicably connected to each other via a program viewing terminal 8 such as a program recording device, and a plurality of program viewing terminals 8 and a general-purpose network 10 such as the Internet. It consists of nine.
- the configuration of the program recommendation system of the third embodiment is the same as the configuration of the program recommendation system shown in FIG. Originally, a form in which one program viewing terminal 8 is connected to the program recommendation server 9 will be described in order to simplify the explanation of the power to which a plurality of program viewing terminals 8 are connected.
- the program viewing terminal 8 receives a program from a broadcast wave or an IP network, views the program, records the program, plays the recorded program, deletes the recorded program, Make a recording reservation.
- Program viewing terminal 8 includes antenna 1701, tuner 1702, TS decoder 17 03, CAS 1704, signal cancel unit 1705, demultiplexer 1706, MPEG video decoder 1707, MPEG Audio decoder 1708, HDMI terminal 1709, recording encryption unit 1 713, recording control unit 1714, recording medium 1715, playback control unit 1716, recording decoding unit 171 7, PSI extraction unit 1718, EPG storage unit 1719, pointer light receiving unit 1720, coordinate extraction unit 1721, cluster identification information extraction unit 1722, cluster storage unit 1723, key 1724, user identification unit 1 725, user information DB 1726, cluster transmission unit 1727, a recommendation receiving unit 1728, a recommended program information extraction unit 1729, a cluster search unit 1730, a recommended reproduction control unit 1732 and an enlargement control unit 1733.
- the program recommendation server 9 includes a cluster reception unit 1750, a user management unit 1751, a user attribute DB1752, a cluster registration unit 1754, a cluster playback time management unit 1755, a cluster DB175 6, a community cluster statistics unit 1757, all users It consists of a cluster statistics unit 1758, a community specialization unit 1759, a recommended cluster creation unit 1760, and a recommendation transmission unit 1761.
- Tuner 1702 receives a broadcast wave of a predetermined channel from a broadcast wave such as terrestrial digital broadcast or BS digital broadcast, and outputs the broadcast wave to TS decoder 1703.
- the TS decoder 1703 demodulates the received general transport 'stream format broadcast wave and outputs it to the descrambling unit 1705.
- CAS 1704 is a general Conditional Access System, reads encryption key information from an IC card in which user registration information and program encryption key are recorded, and outputs the read information to descrambling unit 1705.
- the sign release unit 1705 decrypts the signified transmission / reception signal demodulated by the TS decoder 1703 based on the encryption key from the CAS1704, and decrypts the encrypted program information. And output to the demultiplexer 1706.
- the demultiplexer 1706 separates information multiplexed as program information.
- the program information at least video information, audio information, and program management information are multiplexed. After these are separated, the video information is sent to the MPEG Video decoder 1707, the audio information is sent to the MPEG Audio decoder 1708, and the program management information is sent to the program information.
- the video information is sent to the MPEG Video decoder 1707
- the audio information is sent to the MPEG Audio decoder 1708
- the program management information is sent to the program information.
- the MPEG Video decoder 1707 decodes the video information from the demultiplexer 1706, converts it into video data, and outputs it to the HDMI terminal 1709.
- the MPEG Audio decoder 1708 demodulates the audio information from the demultiplexer 1706, converts it into audio data, and outputs it to the HDMI terminal 1709.
- HDMI terminal 1709 is an HDMI cape
- the HDMI terminal 1710 is connected to the monitor 1711 and the speaker 1712.
- HDMI terminals 1709 and 1710 are HDMI interfaces for general video information and audio information, and mutually authenticate the program viewing terminal 8, the monitor 1711, and the speaker 1712, and share a common encryption key.
- the video data and audio data are encrypted with a common encryption key and output to the monitor 1711 and the speaker 1712.
- the monitor 1711 is a general program video information display device, which decrypts the encrypted video data received at the HDMI terminal 17010 and outputs the video data of the program.
- the SPICE 1712 is a general program audio information output device that decrypts encrypted video data received at the HDMI terminal 1710 and outputs program audio data.
- the recording encryption unit 1713 performs a signal processing for recording the received program information on the recording medium.
- the recording encryption unit 1713 encrypts the program information input from the demultiplexer 1706 and outputs the encrypted program information to the recording control unit 1714. Therefore, the recording encryption unit 1713 performs encryption processing specific to the recording medium for recording the program information.
- DVD-RAM or CPRM (Content Protection for Rewritable Media), Blu-ray Icesk or HD-DVD uses AACS (Advanced Access Content System).
- the recording control unit 1714 controls the recording of the program information encrypted by the recording encryption unit 1713 on the recording medium 1715.
- the recording control unit 1714 records program information on the recording medium by performing modulation processing and error correction coding processing specific to the recording medium.
- Recording medium 1715 is a computer-readable recording medium for recording program information.
- a magnetic recording disk such as a hard disk, an optical disk such as a DVD or Blu-ray disk, or a semiconductor memory such as an SD card.
- magneto-optical recording media such as MO / MD are used.
- the playback control unit 1716 reads the program information recorded on the recording medium 1715 from the recording medium 1715.
- the recording control unit 1714, the recording medium 1715, and the reproduction control unit 1716 are generally configured as a drive system.
- the recording decryption unit 1717 performs decryption corresponding to the encryption of the recording encryption unit 1713. .
- the recording decryption unit 1717 decrypts the encrypted program information signal recorded on the recording medium 1715 and outputs the decrypted signal to the TS decoder 1703. Note that the process of reproducing the program information output to the TS decoder 1703 is the same as the process of reproducing the program received by the antenna 1701 described above.
- the PSI extraction unit 1718 extracts PSI (Program Specific Information) from the program management information extracted by the demultiplexer 1706.
- PSI mainly includes information such as PMT (Program Map Table) and EIT (Event Information Table) TOT (Time Offset Table).
- the PMT stores packet identification information such as images and sounds included in the program! Therefore, if packet identification information such as images and sounds can be obtained from the PMT, the program can be reproduced by extracting packets with the packet identification information.
- the EIT includes information related to the program such as the program name, broadcast date and time, and broadcast content.
- EPG Electronic Program Guide
- the EPG storage unit 1719 creates and stores EPG information using the EIT in the PSI extracted by the PSI extraction unit 1718.
- the pointer light receiving unit 1720 extracts a position indicated by the laser pointer.
- the pointer light receiving unit 1720 is not necessary.
- the coordinate extraction unit 1721 converts the position information irradiated with the laser pointer by the pointer light receiving unit 1720 into coordinate information in the screen.
- the coordinate extraction unit 1721 obtains two-dimensional coordinate information in which the horizontal direction in the screen of the monitor 1711 is the X axis and the vertical direction is the y axis.
- the cluster identification information extraction unit 1722 extracts cluster information from the program management information multiplexed in the program data separated by the demultiplexer 1706.
- the cluster information includes cluster identification information of clusters defined in the program and coordinate information for each cluster.
- a cluster is partial information of a program divided by a user so that it can be selected by two-dimensional coordinates in a video frame.
- a cluster that includes objects such as cast actors in a frame is defined. The same object that appears in consecutive frames on the time axis is given the same cluster identification information.
- it is defined by changing the coordinates in synchronism with the location of the object for each program frame.
- the cluster in this embodiment is rectangular.
- the cluster coordinate information includes coordinates corresponding to the pixel at the upper left corner of the cluster and coordinates corresponding to the pixel at the lower right corner.
- the cluster identification information extraction unit 1722 extracts the coordinate information of the program selected by the user with the pointing device, and compares the coordinate information with the cluster information multiplexed in the program. The cluster selected by the user is identified, and the cluster identification information is extracted and output to the cluster storage unit 1723.
- the cluster information has been described as being multiplexed in the program, the present invention is not limited to this.
- the cluster information may be obtained from a cluster information providing server connected to the general-purpose network 10.
- the program viewing terminal 8 transmits the program identification information of the program viewed and selected by the user to the cluster information providing server, and obtains cluster information corresponding to the program viewed and selected by the user.
- the program identification information of the target program, the time information of the program, and the coordinate information of the selected cluster are transmitted to the cluster information providing server, and the cluster corresponding to the selected cluster is sent. You can get the information.
- the cluster storage unit 1723 includes cluster identification information for identifying a cluster selected by the user using the pointing device extracted by the cluster identification information extraction unit 1722, and the cluster extracted by the PSI extraction unit 1718.
- the time information of the program at the time of selection and the coordinate information of the cluster for each frame are stored in association with the program identification information of the program. Even when the user selects a plurality of clusters, the cluster storage unit 1723 stores a plurality of pieces of cluster identification information.
- a key 1724 is a part that receives input from the user, and includes a remote control and a front panel.
- the user information DB 1726 is a database in which user information of all users who use the program viewing terminal 8 is registered.
- the user identification unit 1725 identifies the user who uses the program viewing terminal 8 based on the user information input using the key 1724 and the user information stored in the user information DB 1726. In addition, when the user identification unit 1725 succeeds in authenticating the user, the user information of the user currently in use is clustered. Output to transmitter 1727.
- the cluster transmission unit 1727 establishes a connection with the program recommendation server 9 connected by the general-purpose network 10, and stores the cluster identification information of the cluster selected by the user stored in the cluster storage unit 1723,
- the user cluster information is obtained by associating the time information of the program at the time of cluster selection, the program identification information for identifying the program, and the coordinate information of the cluster for each frame with the user identification information from the user identification unit 1725 To the program recommendation server 9.
- the recommendation receiving unit 1728 receives recommendation information related to a program recommended by the program recommendation server 9 via the general-purpose network 10.
- the recommended program information extracting unit 1729 extracts recommended program information from the recommended information received by the recommended receiving unit 1728.
- the recommended program information is information that can uniquely identify a recommended program, and is, for example, program identification information, program broadcast date and time, and channel information.
- the recommended program information extraction unit 1729 also includes cluster identification information of a recommended cluster in the recommended program given to the recommended program information, cluster time information indicating the time in the program of the recommended cluster, and coordinates of the recommended cluster. Extract information.
- the cluster search unit 1730 searches the recording medium 1715 based on the program identification information of the recommended program extracted by the recommended program information extraction unit 1729, and the program is recorded! Determine whether or not. If it is determined that the program is recorded, the cluster search unit 173 0 uses the cluster identification information of the recommended cluster recommended by the program recommendation server 9 or the start time information of the recommended cluster on the recording medium 1715. Determine if it is recorded. When it is determined that the recommended cluster is recorded on the recording medium 1715, the cluster search unit 1730 outputs the recommended cluster information to the recommended reproduction control unit 1732.
- the recommended playback control unit 1732 Based on the recommended cluster information input from the cluster search unit 1730, the recommended playback control unit 1732 also plays back the recommended program with the playback start position of the recommended cluster. In addition, the recommended playback control unit 1732 calls the first frame screen of the recommended cluster from the recording medium 1715 and displays thumbnails to prompt the user to play before playback. In this case, the recommended playback control unit 1732 calls a recommended cluster from a plurality of recommended cluster information, creates a thumbnail image, and performs list display. When a thumbnail image is selected by the user, the recommended playback control unit 1732 replays the recommended cluster corresponding to the thumbnail image. Play the program from the live start position.
- the enlargement control unit 1733 enlarges and displays the recommended clusters in the program in accordance with the setting operation from the user.
- the cluster coordinate information is added to the recommended cluster information from the program recommendation server 9, and the recommended cluster of the program is enlarged and displayed based on this coordinate information.
- the enlargement control unit 1733 enlarges and displays the coordinate range indicated by the recommended cluster information with respect to the output from the MPEG Video decoder 1707.
- the enlargement control unit 1733 enlarges the recommended cluster so that it is maximized within the screen.
- the cluster receiving unit 1750 includes a plurality of program viewing terminals connected to the general-purpose network 10.
- the program identification information is program identification information that can uniquely identify the selected program, or program broadcast date and time, channel information, and broadcast station identification information of the selected program.
- the received user cluster information is output to the user management unit 1751.
- User attribute DB1752 includes registered user identification information and, if necessary, user name, gender, address / phone number, manufacturer / model of program viewing terminal, resolution of connected monitor and program viewing terminal. It manages user attribute information such as network IDs as a user registration list. The user performs user registration processing for program recommendation on the program recommendation server 9 in advance. In the user registration process, after user registration information such as user name setting and address / phone number notification is given, the user information is generated and notified to the user on the program recommendation server 9 side. Attribute Stores / manages user attribute information notified by the user in association with user information generated by DB1752.
- the user management unit 1751 determines whether or not the user is a registered user of the user program recommendation server 9 specified by the received user information.
- the user management unit 1751 Search the user registration list stored in DB1752 for user information that matches the received user information. If it is determined that the received user information is registered in the user attribute DB1752, it is determined that the user information is a registered user of the program recommendation server 9, and the processing is continued.
- the cluster registration unit 1754 receives the user cluster information for each user received by the cluster reception unit 1750.
- the cluster DB17
- the cluster registration unit 1754 identifies the cluster identification information of the selected cluster included in the user cluster information, the time position information in the program of the cluster, and the program that includes the cluster.
- the program identification information is extracted and stored in the cluster reproduction time management unit 1755.
- the cluster playback time management unit 1755 stores / manages cluster identification information, time position information, and program identification information in association with each other. Note that if the user cluster information includes the coordinate information on the screen of the cluster in each frame associated with the time position information, the cluster playback time management unit 1755 stores the cluster identification information, the time position information, and The coordinate information is stored.
- the community cluster statistics unit 1757 selects users having the same cluster identification information as the cluster identification information included in the user cluster information received by the cluster reception unit 1750 from the cluster DB 1756 and groups them. Users with the same cluster identification information are grouped based on the cluster identification information managed in the cluster DB 1756 in association with user information. In addition, the community cluster statistics unit 1757 aggregates all clusters other than the cluster indicated by the cluster identification information received by the cluster reception unit 1 750 based on the cluster identification information of all users belonging to the grouped group. The result is normalized by the total number of clusters selected by the users in the group.
- the all-user cluster statistics unit 1758 is registered in the user attribute DB1752 unlike the community cluster statistics unit 1757, which converts all users with cluster identification information received by the cluster reception unit 1750 into a community gnole. All users, that is, all registered users in the program recommendation server 9 are grouped as all user groups including community groups.
- the all-user cluster statistics unit 1758 after grouping all users, selects all clusters other than the cluster indicated by the cluster identification information received by the cluster reception unit 1750 based on the cluster identification information of all users. Aggregate, and normalize the aggregation result with the total number of clusters selected by the registered user of the program recommendation server 9.
- the community specialization unit 1759 subtracts the aggregated results normalized by the all-user cluster statistical unit 1758 from the aggregated results normalized by the community cluster statistical unit 1757, and then creates a cluster specialized for the community group. Create aggregate rankings within community groups. The total ranking within the created community group is output to the recommended cluster creation unit 1760.
- the recommended cluster creation unit 1760 determines the upper cluster of the aggregate ranking in the community group from the community specialization unit 1759, or the frequency of the aggregate ranking as a threshold value, and extracts the cluster having the frequency equal to or higher than the threshold as the recommended cluster . Based on the cluster identification information corresponding to the recommended cluster, the recommended cluster creation unit 1760 receives from the cluster playback time management unit 1755 time information indicating the appearance time (playback time) of the recommended cluster in the program and the time information. Corresponding to the coordinate information of the recommended cluster for each frame. The recommended cluster creation unit 1760 includes cluster identification information for identifying recommended clusters, time information of recommended clusters, coordinate information of recommended clusters, and program identification information for identifying programs including recommended clusters. As a set, recommended program information is created.
- the recommended cluster creation unit 1760 inputs the normalized aggregation results of all user clusters from the all-user cluster statistics unit 1758, ranks the aggregation results in descending order of the number of clusters, and ranks high in the ranking.
- a cluster having the number of clusters equal to or greater than the threshold is extracted as a popular cluster by judging the threshold of the cluster or the aggregation result.
- Recommended cluster generator 176 0 is based on the cluster identification information corresponding to the popular cluster, and from the cluster playback time management unit 1755, the time information indicating the appearance time (playback time) of the popular cluster in the program and one frame corresponding to the time information. Call the coordinate information of each popular cluster.
- the recommended cluster creation unit 1760 includes cluster identification information for identifying popular clusters, time information in programs of popular clusters, coordinate information of popular clusters, and program identification information for identifying programs including popular clusters. As a set, popular program information is created. The recommended program information and popular program information created by the recommended cluster creation unit 1760 are output to the recommendation transmission unit 1761 as recommendation information.
- the recommendation transmission unit 1761 establishes a connection with the program viewing terminal 8 connected via the general-purpose network 10, and displays recommended information including the recommended program information and popular program information created by the recommended cluster creation unit 1760. Send to viewing terminal 8.
- the user A selects a cluster in the program A by the program viewing terminal 8 owned by the user A.
- the program viewing terminal 8 includes user information for identifying user A, cluster identification information for identifying clusters multiplexed in program A, cluster appearance time information, and cluster coordinate information for each frame.
- program identification information for identifying a program including a cluster is transmitted to the program recommendation server 9 as user cluster information.
- the cluster receiving unit 1750 of the program recommendation server 9 receives from the program viewing terminal 8 owned by the user A a user cluster including user information, cluster identification information, time information, coordinate information, and program identification information. Receive information.
- the user management unit 1751 determines whether or not the received user information is stored in the user attribute DB 1752. If the user information is stored, the user management unit 1751 determines that user A is a registered user of the program recommendation server 9, and continues the process. .
- Cluster registration unit 1754 registers cluster identification information, time information, coordinate information, and program identification information in cluster DB 1756 in association with user information for identifying user A.
- the community cluster statistics unit 1757 extracts all users having the same cluster identification information as the cluster identification information included in the received user cluster information from the cluster DB 1756 to form a community group.
- the community cluster statistics unit 1757 statistically processes clusters included in programs other than program A in the community group. To do.
- the community cluster statistics unit 1757 forms ranks in the order of the number of clusters selected from the clusters included in the program other than program A in the community group, and the formed ranking is used as the community specialization unit. Output to 1759.
- the all user cluster statistics unit 1758 forms all users registered in the user attribute DB 1752 of the program recommendation server 9 as an all user group including the community loop formed by the community cluster statistics unit 1757.
- the all-user cluster statistics unit 1758 extracts cluster identification information associated with user information in all user groups from the cluster DB 1756, and statistically processes clusters included in programs other than program A. Then, the all-user cluster statistics unit 1758 forms a ranking in the order of the number of selected clusters among the clusters included in the program other than program A in all user groups, and the formed ranking is Output to the specialization unit 1759 and the recommended cluster creation unit 1760.
- the community specializing unit 1759 subtracts the cluster ranking created by the all-user cluster statistical unit 1758 from the cluster ranking created by the community cluster statistical unit 1757. As a result, the community specializing unit 1759 creates ranking information specialized for the community group of the user who has selected a predetermined cluster in the program A, and outputs it to the recommended cluster creating unit 1760.
- the recommended cluster creation unit 1760 has more clusters than the threshold! /, Clusters, or ranking based on the ranking information of the selected cluster specialized for the community group created by the community specialization unit 1759. A predetermined number of upper clusters are extracted as recommended clusters. Then, the recommended cluster creation unit 1760 creates recommended program information including cluster identification information, program identification information, time information, and coordinate information corresponding to the recommended cluster.
- the recommended cluster creation unit 1760 based on the ranking information of the selected clusters of all registered users created by the all-user cluster statistics unit 1758, is a cluster having a cluster number higher than the threshold or a predetermined higher ranking ranking. A number of clusters are extracted as popular clusters. Then, the recommended cluster creation unit 1760 creates popular program information including cluster identification information, program identification information, time information, and coordinate information corresponding to the popular cluster.
- the recommendation transmission unit 1761 includes recommended program information and people created by the recommendation cluster creation unit 1760. The program information is transmitted to user A's program viewing terminal 8.
- FIGS. 26 to 29 are flowcharts showing the processing flow of the program recommendation system according to the third embodiment, and FIG. 26 transmits user cluster information in the program viewing terminal according to the third embodiment.
- FIG. 27 is a flowchart showing the flow of processing in the program recommendation server according to the third embodiment, and FIG. 28 and FIG. 29 are the programs according to the third embodiment.
- 5 is a flowchart showing a flow of processing after receiving recommendation information in a viewing terminal.
- the user identification unit 1725 displays a user authentication screen for user authentication for the user on the monitor 1711, and the key 1724 accepts input of user information by the user and sends it to the program viewing terminal 8. Is accepted (step S101).
- the user identification unit 1725 determines whether or not the user authentication is successful, that is, whether or not the entered user information matches the user information registered in advance in the user information DB 1726 (Step S1). S 102).
- step S102 If the user authentication fails as a result of the user authentication (NO in step S102), the user identification unit 1725 displays an error message on the monitor 1711 and ends the subsequent processing (step S103). .
- step S104 playback control unit 1716 determines whether or not a playback instruction operation has been performed on a program already recorded on recording medium 1715 (step S104).
- the playback control unit 1716 waits until there is a playback instruction operation of the recorded program from the user.
- step S104 If it is determined that a playback instruction operation has been performed by the user (YES in step S104), the playback control unit 1716 plays the program selected by the user from the recording medium 1715 and outputs the program to the monitor 1711. Playback processing is executed (step S105).
- the demultiplexer 1706 separates the program management information multiplexed in the program
- the PSI extraction unit 1718 extracts the PSI from the program management information
- the cluster identification information extraction unit 1722 The cluster information is extracted from the program management information.
- Cluster information is a single video frame of a program. Includes the cluster identification information of the clusters in the frame and their coordinate information.
- the coordinate extraction unit 1721 determines whether the user has selected a cluster in the program using the pointing device for the program being played back (step S106). That is, when the pointer light receiving unit 1720 receives light from the laser pointer, the coordinate extraction unit 1721 determines that a cluster in the program has been selected. If it is determined that the cluster is selected by the pointing device (YES in step S106), the coordinate extracting unit 1721 extracts the coordinate information in the screen selected by the pointing device (step S107).
- the cluster identification information extraction unit 1722 extracts the cluster identification information of the cluster corresponding to the coordinate information extracted by the coordinate extraction unit 1721 (step
- the extracted cluster identification information is stored in the cluster storage unit 1723 in association with the program identification information of the program together with time information and coordinate information in the program multiplexed in the program of the cluster.
- step S106 After the cluster identification information is extracted and stored in the cluster storage unit 1723, or when it is determined that the cluster is not selected (NO in step S106), the playback control unit 1716 plays the program. Continue processing to determine whether the program being played has ended (step S106).
- step S109 If it is determined that the program being played has not ended (NO in step S109), the process returns to step S105, and playback control unit 1716 continues to execute the program playback process.
- cluster transmission unit 1727 identifies the cluster selected by the user stored in cluster storage unit 1723. Cluster identification information to be recorded, program time information at the time of cluster selection extracted by the PSI extraction unit 1718, cluster coordinate information for each frame, program identification information of a program including a cluster, and user identification unit 1725 The user information of the identified user is transmitted to the program recommendation server 9 as user cluster information (step S110).
- the cluster receiving unit 1750 of the program recommendation server 9 is a program viewing terminal.
- the user cluster information transmitted by 8 is received (step S111).
- the user management unit 1751 receives the user class received by the cluster reception unit 1750. It is determined whether the user information included in the star information matches the user information registered in advance in the user attribute DB 1752 of the program recommendation server 9 (step S112). As a result of user authentication, if it is determined that the user is not a registered user (NO in step S112), the user management unit 1751 creates an error message to that effect and sends the error message to the program viewing terminal 8. Thereafter, the subsequent processing is terminated (step S113).
- the cluster registration unit 1754 receives the cluster identification included in the user cluster information received by the cluster reception unit 1750.
- the information is registered in the cluster DB 1756 in association with the user information included in the user cluster information (step S114).
- the cluster registration unit 1754 registers the cluster identification information included in the received user cluster information, the time information in the program of the cluster, and the coordinate information of the cluster for each video frame in the cluster playback time management unit 1755. To do.
- the community cluster statistics unit 1757 selects all the users having the currently registered cluster identification information from the cluster DB 1756. And the extracted users are grouped into community groups (step S115).
- the community cluster statistical unit 1757 stores the programs other than the program indicated by the user cluster information received by the cluster receiving unit 1750, stored in correspondence with the user information of all the grouped users.
- the cluster identification information in is collected (step S116).
- the community cluster statistics unit 1757 counts the cluster identification information stored for each user.
- the community cluster statistics unit 1757 normalizes the tabulated results totaled in the community group with the total number of clusters of all users in the community group (step S117). As a result, the probability of whether or not the user has selected for each cluster can be calculated, and ranking information indicating the selection probability for the cluster is generated.
- the all-user cluster statistics unit 1758 extracts the users registered in the user attribute DB1752, that is, all users registered in the program recommendation server 9, and is the same as the received cluster identification information.
- Community group consisting of users with cluster identification information Are grouped as all user groups including the group (step S 118).
- the all-user cluster statistics unit 1758 stores a cluster in a program other than the program indicated by the user cluster information received by the cluster receiving unit 1750, stored in correspondence with the user information of all registered users.
- the identification information is totaled (step S119).
- the all-user cluster statistics unit 1758 normalizes the aggregated result of the aggregated cluster identification information of all registered users with the total number of clusters of all registered users (step S 120). As a result, it is possible to generate a ranking in which the selection rate of the cluster of users in all user groups is calculated for each cluster.
- the community specializing unit 1759 subtracts the aggregated cluster identification information in all user groups from the aggregated cluster identification information in the community group. Aggregate clusters specialized for community groups. Then, the recommended cluster creation unit 1760 ranks the subtraction results in descending order of the number of clusters, sets a predetermined number of clusters higher in the ranking as recommended clusters, cluster identification information for identifying the recommended clusters, and the recommended cluster. Recommended program information including time information in the program, coordinate information of the recommended cluster for each moving image frame of the program, and program identification information for identifying the program is created (step S 121).
- the recommended cluster creation unit 1760 extracts time information and coordinate information corresponding to the cluster identification information of the recommended cluster from the cluster playback time management unit 1755, and creates recommended program information. Also, the recommended cluster creation unit 1760 may determine a threshold value for the subtraction result, and a cluster that exceeds the threshold value may be a recommended cluster! /.
- the recommended cluster creation unit 1760 ranks the clustering result of normalized cluster identification information in all user groups created by the all user cluster statistics unit 1758 in descending order of the number of clusters! A predetermined number of clusters at the top of the ranking are regarded as popular clusters, cluster identification information for identifying popular clusters, time information within programs of popular clusters, and coordinate information of popular clusters for each video frame of the program, Popular program information including program identification information for identifying a program is created (step S122).
- the recommended cluster creation unit 1760 determines the threshold value of the result of aggregation by the all-user cluster statistics unit 1758, and a cluster exceeding the threshold value may be regarded as a popular cluster! /.
- the recommendation transmitting unit 1761 transmits the recommended program information and popular program information created by the recommended cluster creating unit 1760 as recommended information to the program viewing terminal 8 (step S 1 23).
- the recommendation receiving unit 1728 of the program viewing terminal 8 receives the recommended information including the recommended program information and the popular program information transmitted by the recommendation transmitting unit 1761 of the program recommendation server 9 ( Step S131).
- the recommended program information extracting unit 1729 determines whether or not recommended program information has been extracted from the recommended information received by the recommended receiving unit 1728 (step S 132).
- the recommended program information extraction unit 1729 prompts the user to select either recommended program information or popular program information.
- recommended program information is selected using the key 1724
- the recommended program information extraction unit 1729 extracts recommended program information.
- popular program information is selected
- the recommended program information extraction unit 1729 extracts popular program information. To do. It should be noted that setting for extracting both recommended program information and popular program information is not a problem. This can provide more options for the user.
- step S132 If it is determined that the recommended program information has been extracted (YES in step S132), the cluster search unit 1730 is based on the program identification information included in the recommended program information extracted by the recommended program information extraction unit 1729. Then, it is determined whether or not a program having the same program identification information as the program identification information is recorded on the recording medium 1715 (step S 133). If it is determined that the program including the recommended cluster is not recorded on the recording medium 1715 (NO in step S133), the process proceeds to step S140.
- cluster search unit 1730 calls program information corresponding to the program identification information from recording medium 1715. Then, the time information and the coordinate information in the program of the recommended cluster included in the received recommended program information are extracted and output to the recommended reproduction control unit 1732.
- the recommended playback control unit 1732 displays a thumbnail image of the recommended cluster on the monitor 1711 based on the time information and the coordinate information extracted by the cluster search unit 1730 (step S134).
- the recommended playback control unit 1732 determines whether or not the user has issued a thumbnail image selection and playback instruction (step S 135).
- playback of thumbnail images If it is determined that an instruction has not been issued (NO in step SI 35), the process proceeds to step S140.
- playback control unit 1716 reads a program corresponding to the program identification information extracted by cluster search unit 1730 from recording medium 1715. Then, the program is played back from the playback start time of the recommended cluster included in the time information extracted by the cluster search unit 1730 (step S136).
- the expansion control unit 1733 determines whether or not an expansion instruction for expanding the recommended cluster by the user has been issued (step S 137).
- the enlargement instruction is issued when the input is accepted by the key 1724. If it is determined that an enlargement instruction has been issued (YES in step S137), the enlargement control unit 1733 receives the recommended cluster enlargement instruction from the user and extracts the video of the recommended cluster extracted by the cluster search unit 1730. Based on the coordinate information in units of frames, a recommended cluster is cut out from the display image of the program and enlarged (step S138).
- step S137 After the recommended cluster is displayed in an enlarged manner, or when it is determined that an enlargement instruction has not been issued (NO in step S137), the playback control unit 1716 plays back the program including the selected recommended cluster. It is determined whether or not is completed (step S139). Note that the playback control unit 1 716 automatically ends the program playback processing when the playback end time of the recommended cluster included in the time information extracted by the cluster search unit 1730 is reached. If it is determined that the program playback process has not ended (NO in step S139), playback control unit 1716 continues the playback process.
- step S139 cluster search unit 1730 displays all recommended clusters included in the recommended program information transmitted by the program recommendation server 9. It is determined whether or not the reproduction process has been completed (step S140). In step S140, cluster search unit 1730 determines whether or not the processing in steps S133 to S139 has been completed for all the recommended clusters included in the received recommended program information.
- step S140 if it is determined that the playback processing of all the recommended clusters has not been completed (NO in step S140), the processing returns to step S133, and the processing for the remaining recommended clusters is performed. Done. On the other hand, if it is determined that the playback process for all the recommended clusters has been completed (YES in step S140), all the processes are terminated.
- the cluster search unit 1730 extracts the recommended program information. Based on the program identification information included in the popular program information extracted by the section 1729, it is determined whether or not a program having the same program identification information as the program identification information is recorded on the recording medium 1715 (step S141). ). If a program including a popular cluster is recorded on the recording medium 1715 and it is determined that it is! /, NA! / (NO in step S141), the process proceeds to step S148.
- cluster search unit 1730 calls program information corresponding to the program identification information from recording medium 1715.
- the time information and the coordinate information in the program of the popular cluster included in the received popular program information are extracted and output to the recommended playback control unit 1732.
- the recommended playback control unit 1732 displays the thumbnail images of popular clusters on the monitor 1711 based on the time information and the coordinate information extracted by the cluster search unit 1730 (step S 142).
- the recommended playback control unit 1732 determines whether or not the user has issued a thumbnail image selection and playback instruction (step S143). If it is determined that a thumbnail image playback instruction has not been issued (NO in step S143), the process proceeds to step S148. On the other hand, if it is determined that a playback instruction has been issued (YES in step S143), playback control unit 1716 reads a program corresponding to the program identification information extracted by cluster search unit 1730 from recording medium 1715. Then, the program is played from the playback start time of the popular cluster included in the time information extracted by the cluster search unit 1730 (step S 144).
- the expansion control unit 1733 determines whether or not an expansion instruction for expanding the popular cluster by the user has been issued (step S145).
- the enlargement instruction is issued when the input is accepted by the key 1724. If it is determined that an enlargement instruction has been issued (YES in step S145), the enlargement control unit 1733 receives the popular cluster enlargement instruction from the user, and receives the popular cluster extracted by the cluster search unit 1730. Based on the coordinate information in units of moving picture frames, a popular cluster is cut out from the display image of the program and enlarged (step S146).
- the playback control unit 1716 plays the program including the selected popular cluster. It is determined whether or not is completed (step S147). Note that the playback control unit 1 716 automatically ends the program playback processing when the playback end time of the popular cluster included in the time information extracted by the cluster search unit 1730 is reached. If it is determined that the program playback process has not ended (NO in step S147), playback control unit 1716 continues the playback process.
- step S147 the cluster search unit 1730 displays all popular clusters included in the popular program information transmitted by the program recommendation server 9. It is determined whether or not the reproduction process has been completed (step S148). In step S148, cluster search unit 1730 determines whether or not the processes in steps S141 to S147 have been completed for all popular clusters included in the received popular program information.
- step S148 if it is determined that the reproduction processing of all popular clusters has not been completed (NO in step S148), the processing returns to step S141, and the processing for the remaining popular clusters is performed. On the other hand, if it is determined that all popular clusters have been replayed (YES in step S148), all processes are terminated.
- the program viewing terminal 8 receives a broadcast program in which the program is spatially divided into clusters, and is selected by the user using a pointing device.
- the cluster is transmitted to the program recommendation server 9.
- the program recommendation server 9 can manage selected clusters for each user, group users according to the selected clusters, and provide recommended programs to the users by collaborative filtering within the groups.
- a group including a community group for example, so that a specific program can be recommended in a community group composed of users who have selected the same cluster, for example, Create a group consisting of all users,
- a group consisting of all users By removing the influence of popular popular programs that are preferred by the users from programs recommended within the community group, it becomes possible to recommend programs that are unique within the community group.
- popular programs can be easily known by users without using such a recommendation system, so it is useful to recommend programs specialized for community groups.
- a method for recommending a generally popular program can be realized by using a known method.
- the power describing a method of transmitting one cluster selected by the user is not limited to this.
- a plurality of user cluster information may be transmitted to the program recommendation server 9 as a cluster selection history from the user.
- users are grouped according to the degree of coincidence between the cluster selection history for each user and the selection history of other users. For example, users having all the clusters included in the transmitted cluster selection history as a selection history are grouped.
- the selection history in this case may be determined based on the similarity of the cluster selection history within one specific program, or may be transmitted when, for example, several histories are stored. In this way, grouping based on the cluster selection history makes it possible to make a group with more consistent preferences, leading to improved accuracy of recommended programs and clusters.
- the above-described form of transmitting the cluster selection history is also applicable to the first and second embodiments described above.
- the program viewing terminal 2 transmits a history of program reservation information
- the program recommendation server 3 transmits a reservation information history similar to the reservation information history of the transmitted program. Group users with Even in this case, similar reservation information may be a complete history match, or users who match more than half of the history information to be sent may be grouped.
- the program viewing terminal 5 transmits a plurality of marked program scene histories to the program recommendation server 6, and the program recommendation server 6 transmits the history of the marked program history. Group users by similarity.
- users can be grouped by comparing history information, so that users with more matched preferences can be grouped. Therefore, it leads to the improvement of the precision of the recommendation information produced
- the method of grouping all users as a group that includes all grouped users in the program recommendation server has been described, but the present invention is not limited to this.
- User attribute information such as gender, age, and residence is managed as user information.
- all user gnoles (second gnole) including community groups (first group) It is also possible to generate recommendation information by subtracting the aggregated results of all user groups from the aggregated results in the community gnole. In this way, even if popular programs are biased due to, for example, gender, age, or address, it is possible to eliminate this effect and create more specialized recommendation information within the community group. Become.
- the reservation rate is normalized by the number of users who have made reservations. For example, for broadcasts of a genre with high broadcast frequency such as a news program, Normalization may be performed by broadcast time.
- the marking weight may be reduced, You can change the marking weight according to the frequency of marking.
- the aggregation when the aggregation within the group is performed, the aggregation may be performed by weighting a specific user whose reliability of the recommendation information is considered to be high. Users whose reliability of recommendation information is considered to be high may be automatically extracted based on the degree of coincidence between the user's reserved program and the recommendation information in the group, or manage other users or program recommendation systems It may be chosen artificially by an administrator. [0386] For example, a specific day such as New Year's Eve or a sports tournament held once every several years may be excluded from the target of counting the reservation rate, or the weight may be changed.
- marking by button press time marking by inputting the start time and end time of the program scene, and the start time and duration of the program scene are designated.
- Marking powers marking powers that specify the end time and duration of a scene, but are not limited to these. For example, you can mark a program scene as a favorite scene start time when you return from a user's favorite scene button press time as a favorite scene start time. You can also use fifty yarn and scene markings!
- the scene change immediately before and after the user presses the favorite scene button is automatically detected, and the scene change is marked as the favorite scene start time and end time respectively. Also good. In this way, the burden on the user marking the favorite scene can be reduced.
- the scene change is determined based on a video-based scene change that determines whether the scene has changed from the change in the histogram extracted for each frame of the program, or determines the change in the performer projected by image recognition. And voice-based scene change judgments that use voice recognition to detect changes in sound tone, volume, or changes in performers using voice recognition.
- weighted addition may be performed.
- the weight may be changed according to the designation of the recording image quality such as the high image quality mode or the long time mode.
- the user A himself / herself selects a community duplication to participate from among a plurality of community groups prepared in advance, and this community group As a whole set, statistics with programs reserved by user A may be calculated. This makes it possible to match the user needs.
- the user A can select a group to which the user A belongs in the community reservation information statistics unit 236 in the first embodiment.
- the program recommendation server can recommend a program having high similarity to the preference of the group selected by the user A and the program reserved by the user A for recording.
- the user can select to join / unsubscribe from the community group. This makes it possible to recommend programs that change according to the user's preferences or that match the user's mood.
- the user may be able to search for community groups. This allows the user to search for a community group that the user wants from a number of community groups.
- the search query in this search is preferably based on the user's search keyword input.
- other search queries such as selection from a category, image information, video information, music information, and voice are used as search queries. Even if the user's gesture is used, the same effect can be obtained.
- a history of community group groups that the user has selected in the past may be displayed. This allows users to easily select community groups that they have selected in the past.
- the ranking of community groups selected by other users may be displayed. Preferably a large number selected by other users! /, It is desirable to display the ranking of community groups! /. This allows the user to select more popular! / Community groups. As another example, the ranking of community groups that have rapidly increased the selectivity of other users may be displayed. As a result, the momentary popularity is increasing. A community group can be selected.
- the user may be able to specify the threshold determination criterion in the recommended program creation unit 239. For example, when the user raises the threshold judgment criterion, it is possible to recommend only programs with a higher reservation rate in the gnolepe. On the other hand, it is possible to recommend more programs by lowering the threshold criteria.
- the program selected by the user may be weighted when the aggregation within the group is performed.
- a mechanism may be considered in which the date and time of a program recorded by the user is stored and the weight decreases as time passes.
- the program viewed by the user without recording but only the programs recorded by the user in the past is used for the aggregation of the group. Also good. As a result, it is possible to increase the number of samples of the program selected by the user, and it is possible to recommend a program that more matches the user's preference.
- the program recommendation server 3 shown in FIG. 4 described in the first embodiment is used.
- the specialization generation unit 30 in FIG. 4 a method for creating a recommended program specialized for a community group to which the user belongs, which is calculated from a user's reserved program information group by calculating a tf ′ idf (term frequency inverse document frequency) value. Will be described.
- tf 'idf value is a method of weighting from the appearance frequency of the reserved program information of the program, tf represents the appearance frequency of specific reserved program information in the target community group, id f is all users
- the reserved program information of a specific program is included in the reserved program information of! /, Indicating the reciprocal of the number of users. If certain reserved program information appears frequently in one community group and the number of users in which the reserved program information appears is small, it is a characteristic of community groups with high tf .idf values. It can be said that it is reserved program information. Therefore, this characteristic reserved program information is generated as a recommended program specialized for the community group.
- the configuration of the fourth embodiment for generating a recommended program using the tf 'idf value is the same as that of the specialization generating unit 30 of the program recommendation server 3 in FIG. Section 236, All User Reservation Information Statistics Section 237, and Community Specialization Section 238.
- the community reservation information statistical unit 236 calculates tf of the tf'idf value, and extracts reserved program information having a high appearance frequency within the community group. Actually, the community reservation information statistics section 236 counts up the reservation program information group held for each user belonging to the community group for each reservation program information one by one in the community group, and the appearance frequency The reserved program information with a high value is extracted, and the appearance frequency is calculated as tf.
- FIG. 30 is a flowchart showing a process flow of the community reservation information statistics unit 236 according to the fourth embodiment.
- the community reservation information statistics unit 236 acquires the reservation program information of all users in the community group to which the user (authenticated user) authenticated by the user management unit 231 belongs from the reservation information DB 235 (step S 161 ).
- the reserved program information acquired here is reserved program information transmitted by a user belonging to the same community group as the user whose user information is authenticated by the user management unit 231, that is, a program reserved in the community group of the authenticated user. Is reserved program information.
- the community reservation information statistics unit 236 displays the community group to which the authenticated user belongs. The total number of all reserved program information in the program is acquired (step S162). Next, the community reservation information statistic section 236 provides information for each reserved program within the community group.
- the community reservation information statistical unit 236 calculates the appearance frequency tf of each reserved program information in the community group to which the authenticated user belongs (step S164). Specifically, the community reservation information statistics unit 236 calculates the appearance frequency tf of reserved program information by dividing the number of reserved program information in the community group by the total number of all reserved program information in the community group. .
- the community reservation information statistics unit 236 determines whether or not the appearance frequency tf of all reservation program information within the community group has been calculated (step S165).
- the appearance frequency tf of all the reserved program information is calculated and determined to be! /, NA! / (NO in step S165)
- the community reservation information statistics unit 236 proceeds to the process of step S163. Return, and count the number of reservations for other reserved program information in the community group.
- the community reservation information statistics unit 236 displays the information on all community groups to which the authenticated user belongs. It is determined whether or not the appearance frequency tf of the reserved program information has been calculated (step S166). That is, if the authenticated user belongs to multiple community groups, the appearance frequency tf of the reserved program information is calculated for all community groups to which the authenticated user belongs.
- step S166 if it is determined that the appearance frequency tf of the reservation program information of all community groups has not been calculated (NO in step S166), the community reservation information statistics unit 236 performs the processing of step S161. Return to the reservation information DB 235 for the reserved program information of all users in other community groups to which the authenticated user belongs. On the other hand, if it is determined that the appearance frequency tf of the reserved program information of all community groups has been calculated (YES in step S166), the process ends.
- All user reservation information statistics section 237 calculates the idf of tf 'idf value, and among all the reserved program information registered in the reservation information DB 235, all reservation program information in the community group Calculate the appearance frequency of each one. Reservation program information in the community group When the appearance frequency is high in the reserved program information of all users, idf becomes small and it is determined that the reserved program information is not represented by the community group. On the other hand, if the frequency of appearance is low in the reserved program information capability S in the community gnope and the reserved program information of all users! /, The idf will increase and the reserved program information represented by the community group will be increased. It is determined to be information.
- the all-user reservation information statistics unit 237 adds a certain reserved program information k! /, Log ((total number of users) ⁇ (number of users having reserved program information k)) + Calculate idf by 1.
- FIG. 31 is a flowchart showing a process flow of the all-user reservation information statistics unit 237 according to the fourth embodiment.
- all users reservation information statistics unit 237 acquires the total number of users registered in the user attribute DB232 through the user management unit 231 and the reservation information registration section 2 34 (step S 171) 0 then The all-user reservation information statistics unit 237 counts the number of users who have made reservations for each reserved program information in the community group of the authenticated user (step S 1 72).
- the all-user reservation information statistics unit 237 calculates the idf of each reserved program information in the community group to which the authenticated user belongs (step S173). Specifically, the all user reservation information statistics unit 237 calculates idf of each reserved program information based on the following equation (1).
- idf log ((total number of users) ⁇ (number of users with each reserved program information)) + 1 ⁇ ⁇ ⁇ (1)
- the all-user reservation information statistics unit 237 determines whether or not idf has been calculated for all reservation program information in the community group (step S174). Here, if it is determined that the idf of all the reserved program information has not been calculated (NO in step S174), the all user reserved information statistics unit 237 returns to the process of step S172, and others in the community group. For the reserved program information, the number of users who have made reservations is counted. On the other hand, if it is determined that the idf of all the reserved program information has been calculated (YES in step S174), the process ends.
- idf is calculated using the total number of users and the number of users having each reserved program information.
- the present invention is not particularly limited to this, and all community groups The idf may be calculated using the number and the number of community groups having each reserved program information.
- the community specialization unit 238 determines all users in the community group.
- the tf ⁇ idf value is calculated for each reserved program information and output to the recommended program creating unit 239.
- the tf 'idf value is calculated by tf X idf.
- the recommended program creating unit 239 extracts, as recommended programs, programs to be recommended that have a high tf'idf value and are not yet reserved by the user who makes the recommendation. Therefore, all users who have a high appearance frequency in the community group are recommended programs that do not have a high appearance frequency, that is, have a large idf, and have not yet been reserved by the recommending user. As a result, the user is recommended a program that is specialized for the community group, that is, that matches the user's preferences! /.
- FIG. 32 is a flowchart showing a process flow of the community specializing unit 238 according to the fourth embodiment.
- the community specialization unit 238, for the reserved program information in the community group to which the authenticated user belongs, the tf acquired by the community reservation information statistics unit 236 and the idf acquired by the all user reservation information statistics unit 237. are used to calculate the tf ⁇ idf value (step S 181).
- the community specializing unit 238 determines whether or not the tf'idf values of all the reserved program information in the community group to which the authenticated user belongs have been calculated (step S182). Here, if the tf ⁇ idf values of all the reserved program information are calculated and determined as! /, NA! / (NO in step S182), the community specializing unit 238 performs the process of step S181. Return to, and calculate the tf.idf value of other reserved program information in the community group to which the authenticated user belongs.
- the community specializing unit 238 determines the recommendation threshold for determining the program to be recommended. Find the value.
- the recommendation threshold is preferably a value that can be easily calculated, such as an average value of tf′idf values in the reserved program information in the community group. Therefore, the community specializing unit 238 calculates an average value of tf • idf values of all the reserved program information in the community group (step S183).
- the community specializing unit 238 extracts reservation program information having a tf'idf value equal to or greater than the recommendation threshold (average value) as recommendation target reserved program information.
- the community specializing unit 238 determines whether or not the recommended user reserved program information is reserved by the authenticated user! /, Nana! /, And reserved program information (step S185). If it is determined that there is no reserved program information that is not reserved by the authenticated user (NO in step S185), the process ends.
- the community specializing unit 238 makes a reservation for which the authenticated user has not reserved. From the program information, a predetermined number of reserved program information is extracted as a recommended program in descending order of tf'idf value, and recommended program information including program identification information for identifying the recommended program is created (step S 186).
- FIG. 33 is a diagram for explaining extraction of a recommended program using the tf′idf value.
- the horizontal axis in the table 301 shown in FIG. 33 represents program Arufa ⁇ zeta reserved, the vertical axis represents the community two Tigunorepu C L ⁇ c4, the user A ⁇ P belonging to each community group .
- Circles in Table 301 indicate which programs the user has reserved.
- the user A is an authenticated user whose user information has been authenticated by the user management unit 231.
- the community specializing unit 238 obtains a tf 'idf value, and recommends a recommended program based on the tf' idf value. Extract.
- the tf 'idf value is calculated for each reserved program information of the community groups cl and c2.
- the community special 238 calculates the tf'idf value in the same manner for the other warps. Then, the community specializing unit 238 calculates an average value of the calculated 10 tf ⁇ idf values, and sets reserved program information having a tf.idf value equal to or greater than the average value as recommended target reserved program information.
- program ⁇ and program ⁇ in community group cl are recommended reserved programs
- program ⁇ , program ⁇ , and program ⁇ in community group c2 are recommended target reserved programs.
- the program ⁇ and the program ⁇ have a higher tf ′ idf value than the average value, but since the program is already reserved by the user A, the program / 3 and the program ⁇ among the remaining programs are It is extracted as a recommended target reserved program.
- the program tf ′ idf value of the program / 3 is larger than the tf ′ idf value of the program ⁇ , it is recommended for the program / 3 power and is recommended for the program ⁇ power third.
- the reserved program information characteristic of the community group to which the authenticated user belongs is extracted for all users registered in the user attribute DB 232 and is not all users.
- the reserved program information may be extracted for users who belong to a predetermined plural community gnole.
- the predetermined plural community groups can be set by the following method, for example.
- the idf value is obtained using the total number of users as described in the present embodiment.
- a community group whose community reservation rate (recommended number of recommended programs in the community group ⁇ number of users in the community group) of the program recommended in the first process is equal to or higher than a predetermined threshold (for example, 0.5) is selected. Neighboring community gnole.
- a community group with a high program reservation rate recommended to an authenticated user can be said to be a community group (neighboring community group) whose program reservation tendency is similar to the community group to which the authenticated user belongs.
- the idf value is obtained using the total number of users in the neighborhood community gnole and the community gnole to which the authenticated user belongs. That is, the idf value of the reserved program information k is given by the following equation (2) Is calculated based on
- idf log ((Number of users in the neighborhood community group + Number of users in the community group to which the authenticated user belongs) ⁇ Number of users in which the reserved program information k appears) + 1 ⁇ ⁇ ⁇ (2)
- program ⁇ is recommended (first place) to authenticated user A! /.
- the community reservation rates of the recommended program / 3 in the community groups c3 and c4 to which the authenticated user ⁇ does not belong are 0.5 and 0.3, respectively.
- Community gnole c3 greater than or equal to the threshold (0.5) is the neighborhood community gnole.
- the neighborhood community group may be updated by periodically performing the same process as the first program recommendation. In this way, even if the program reservation tendency of the community group changes with the passage of time, program recommendation specialized for the community group can be performed.
- this neighborhood community gnole can be judged as a community gnole with relatively close preference. For this reason, it is possible to provide a mechanism for recommending the user to join a nearby community group. As a result, it can be expected that the user can easily find a community group close to his / her preference.
- the force S using the average value of tf'idf values as a program recommendation threshold is not limited to this.
- the synergistic average or median value is used as the recommendation threshold value, so that the program recommendation that the user can convince can be performed with high accuracy S.
- the community specialization unit 238 compares the recommended reserved program information with the reserved program information reserved by the authenticated user! /, Ruru community reservation information statistics unit 236 and The all-user reservation information statistics unit 237 may compare the reserved program information to be processed with the reserved program information of the authenticated user and exclude the reserved program information of the authenticated user from the processing target. With this configuration, the processing target reserved program information in the community reservation information statistics unit 236 and the all user reservation information statistics unit 237 is reduced, and the calculation processing amount can be reduced.
- the tf'idf value may be calculated using information other than the reserved program information of the program.
- the tf 'idf value is calculated from the title of the content created by the user, such as a blog or diary, or a word that appears in the content. You can recommend a program whose word is included in the title or detailed information of the program information (information obtained from EPG, tags associated with the program, etc.).
- program information and words represented by a community group may be associated as detailed information of the community group. By doing so, it can be used as tag information when searching for community groups, or when a community group name is displayed, a list is output at the same time, so that when a user selects a community group, The effect is to make it easier to select community groups that match user preferences.
- a program recommendation system is connected to a program viewing terminal for viewing a received program, and the program viewing terminal so as to be communicable with each other via a network, and a user operation on the program
- a program recommendation system comprising: a program recommendation server that obtains operation information representing the information from the program viewing terminal and creates recommendation information related to a program that matches the user's preference according to the operation information.
- the viewing terminal includes an operation receiving unit that receives an input of the operation information, and an operation information transmitting unit that transmits the operation information received by the operation receiving unit to the program recommendation server.
- the recommendation server includes an operation information receiving unit that receives the operation information transmitted by the operation information transmitting unit, an operation information storage unit that stores the operation information received by the operation information receiving unit for each user, A first grouping the users into a first group based on the operation information received by the operation information receiving unit and the operation information for each user stored in the operation information storage unit! A grouping unit; a second grouping unit that groups the users into a second group that includes the first group grouped by the first grouping unit; and the grouping unit configured by the first grouping unit. The grouping result of the operation information of the users belonging to the first group and the second grouping unit.
- a recommendation information generation unit that generates the recommendation information unique to the first group based on the totaled result of the operation information of the users belonging to the second group, and the recommendation information generation unit
- a recommendation information transmission unit that transmits the recommendation information generated by the program viewing terminal to the program viewing terminal, and the program viewing terminal receives a recommendation information transmission unit that receives the recommendation information transmitted by the recommendation information transmission unit.
- a program control unit that performs at least one of selection, recording, reproduction, and deletion of a program based on the recommendation information received by the recommendation information receiving unit.
- the program viewing terminal accepts input of operation information indicating the user's operation on the program, and the received operation information is transmitted to the program recommendation server. Then, in the program recommendation server, the operation information transmitted by the program viewing terminal is received, and the received operation information is stored for each user in the operation information storage unit. Next, based on the received operation information and the operation information for each user stored in the operation information storage unit, the users are grouped into the first group and the second group including the first group. Users are grouped into groups. Based on the totaled result of operation information of users belonging to the first group and the totaled result of operation information of users belonging to the second group, recommendation information specific to the first group is generated and generated. The recommended information is sent to the program viewing terminal.
- the recommendation information transmitted by the program recommendation server is received, and at least one of program selection, recording, reproduction, and deletion is performed based on the received recommendation information.
- a program operated by another user having the same preference as that of the first user is extracted based on the total result of the operation information of the users belonging to the first group, and the first group including the first group is extracted.
- the programs that are generally preferred are extracted, and the programs extracted based on the first group are extracted based on the second group. Since the influence of the selected program is eliminated, it is possible to eliminate programs with a high degree of preference in general, and to recommend programs that are specific to specific user preferences.
- the program viewing terminal further includes a marking receiving unit that receives a marking of a reproduction position in the program by the user, and the operation information includes the marking receiving unit. It is preferable to include marking time information indicating a time corresponding to a reproduction position in the program marked by.
- the marking of the reproduction position in the program by the user is accepted, and the operation information includes the marking time information indicating the time corresponding to the reproduction position in the marked program. Therefore, the user's preference can be determined based on the reproduction position in the marked program.
- the program is broadcast in a state in which a display screen is divided into a plurality of areas, and the program viewing terminal has the plurality of programs on the display screen.
- An in-screen selection receiving unit for receiving selection by the user of at least one of the regions, and the operation information includes time information in the program at the time when the selection is received by the in-screen selection receiving unit. It is preferable.
- the program is broadcast in a state where the display screen is divided into a plurality of regions, and at the program viewing terminal, at least one of the plurality of regions on the display screen is selected by the user. Accepted.
- the operation information includes time information in the program at the time when the selection is accepted. Therefore, the user's preference can be determined based on at least one area selected by the user among the plurality of areas on the display screen.
- the program recommended by the program recommendation server can be automatically reproduced from the reproduction position corresponding to the time information.
- the program is broadcast in a state in which a display screen is divided into a plurality of areas, and the program viewing terminal has the plurality of programs on the display screen.
- An in-screen selection receiving unit that accepts selection by the user of at least one of the regions is further included, and the operation information includes position information on the display screen from which selection has been accepted by the in-screen selection receiving unit. It is preferable.
- the program is broadcast in a state where the display screen is divided into a plurality of regions, and at the program viewing terminal, at least one of the plurality of regions on the display screen is selected by the user. Accepted.
- the operation information includes position information on the display screen where the selection is accepted. Therefore, the user's preference can be determined based on at least one area selected by the user from among the plurality of areas on the display screen. Further, only a predetermined area in the program recommended by the program recommendation server can be enlarged and reproduced.
- the recommendation information includes area identification information for identifying the area selected by the user from among the plurality of areas. That's right. According to this configuration, since the recommendation information includes the area identification information for identifying the area selected by the user from among the plurality of areas, the area in the program is specified based on the area identification information. The specified area can be reproduced.
- the recommendation information includes reproduction start time information related to the reproduction start position of the recommended program.
- the recommended information includes the reproduction start time information related to the reproduction start position of the recommended program. Therefore, even in the program in which the program recommended by the program recommendation server is not reproduced from the beginning. Can be reproduced from the reproduction start position according to the user's preference.
- the recommendation information preferably includes packet identification information for identifying a specific packet among a plurality of packets constituting a recommended program. .
- the recommendation information includes packet identification information for identifying a specific packet among a plurality of packets constituting the recommended program, a program recommended by the program recommendation server is displayed. Even in programs that do not play from the beginning, user preferences It is possible to reproduce from a specific packet according to the preference.
- the first grouping unit extracts a user group having operation information equivalent to the operation information received by the operation information receiving unit from the operation information storage unit.
- the user group is preferably dubbed as the first group.
- the user group having the operation information equivalent to the received operation information is extracted from the operation information storage unit, and the extracted user group is grouped as the first group. Therefore, users with similar operation information can be grouped as the first group to group users with similar preferences.
- the operation information includes program identification information for identifying a program that is a user's operation target
- the first grouping unit includes: A user group having the same program identification information as the program identification information included in the operation information received by the operation information receiving unit is extracted from the operation information storage unit, and the user group is extracted from the first group. It is preferable to group them as follows.
- the operation information includes the program identification information for identifying the program that is the operation target of the user, and the same program as the program identification information included in the received operation information
- a group of users having identification information is extracted from the operation information storage unit, and the extracted user group is grouped as a first group. Therefore, by grouping a group of users having the same program identification information, users with similar preferences can be grooved.
- the first grouping unit has a time overlap with the marking time information included in the operation information received by the operation information receiving unit. It is preferable to extract a group of users having marking time information from the operation information storage unit and group the group of users as the first group.
- a group of users having marking time information that overlaps with the marking time information included in the received operation information is extracted from the operation information storage unit.
- the extracted user group is grouped as a first group. Therefore, it is possible to group users who have similar preferences by grouping a group of users who have marked overlapping portions in the program.
- the first grouping unit includes a group of users having the same region identification information as the region identification information included in the operation information received by the operation information receiving unit. Are extracted from the operation information storage unit, and the user group is preferably grouped as the first group.
- the user group having the same area identification information as the area identification information included in the received operation information is extracted from the operation information storage unit, and the extracted user group is the first Grouped as a group. Therefore, by grouping a group of users having the same region identification information, it is possible to group users having similar preferences with the force S.
- the second grouping unit groups all users registered in advance in the program recommendation server as the second group.
- the second grouping unit is pre-registered in the attribute information input in advance by a user registered in advance in the program recommendation server or in the program recommendation server. It is preferable to group a plurality of users having attribute information equivalent to attribute information extracted from existing users as the second group.
- the attribute information equivalent to the attribute information input in advance by the user registered in advance in the program recommendation server or the attribute information extracted from the user registered in advance in the program recommendation server A number of users with a group are grouped as the second group. Therefore, multiple users with equivalent attribute information Therefore, the number of users belonging to the second group can be reduced, and the time required for the process of generating the recommended program can be reduced.
- the recommendation information generation unit normalizes the totaled result of the operation information included in the first group by the number of users in the first group, and The operation results included in group 2 are normalized by the number of users in the second loop, and the results of the second group are normalized from the normalized results of the first group. It is preferable that the recommendation information peculiar to the first gnole is generated based on the subtraction result.
- the operation information aggregation result included in the first group is normalized by the number of users in the first group, and the operation information aggregation result included in the second group is the first result. Normalized by the number of users in the two groups, the normalized result of the second group is subtracted from the normalized result of the first group, and a recommendation specific to the first group based on the subtraction result Information is generated.
- the recommendation information generation unit calculates tf (Term Frequency) of the operation information included in the first group, and is included in the second group.
- An idf (Inverse Document Frequency) of the operation information is calculated, a tf ′ idf value for each operation information is calculated based on the tf and the idf, and based on a tf ′ idf value for each operation information,
- the recommendation information specific to the first group is generated.
- tf Term Frequency
- idf Inverse Document Frequency
- tf And tf 'idf value for each operation information is calculated based on and idf.
- recommendation information specific to the first group is generated based on the tf′idf value for each operation information.
- the recommendation information specific to the first group is set. Since the information is generated, it is possible to eliminate programs that are generally highly preferred and recommend programs that are specific to specific user preferences.
- a program viewing terminal represents a program receiving unit that receives a program included in a broadcast wave or communication data, and a user operation on the program received by the program receiving unit.
- the operation accepting unit that accepts input of the operation information and the operation information accepted by the operation accepting unit are communicably connected via a network, and the recommended information regarding the program that matches the user's preference is the operation.
- An operation information transmitting unit that transmits to a program recommendation server created according to information, a recommendation information receiving unit that receives the recommendation information transmitted by the program recommendation server, and the recommendation information receiving unit that receives the recommendation information
- a program control unit that performs at least one of program selection, recording, reproduction, and deletion based on the recommendation information.
- a program viewing program includes a program receiving unit that receives a program included in a broadcast wave or communication data, and a user operation on the program received by the program receiving unit.
- the operation reception unit that receives the input of the operation information representing the information, and the operation information received by the operation reception unit are connected to be communicable via a network, and the recommended information regarding the program that matches the user's preference.
- the operation information transmitting unit that transmits to the program recommendation server created according to the operation information, the recommendation information receiving unit that receives the recommendation information transmitted by the program recommendation server, and the recommendation information receiving unit. Based on the recommendation information, the computer is caused to function as a program control unit that performs at least one of program selection, recording, reproduction, and deletion.
- a program viewing method includes a program receiving step of receiving a program included in a broadcast wave or communication data, and the program received in the program receiving step!
- the operation accepting step for accepting input of the operation information representing the user's operation and the operation information accepted in the operation accepting step are communicably connected via a network and meet the user's preference.
- An operation information transmission step for transmitting recommended information related to a program to a program recommendation server created according to the operation information, a recommendation information reception step for receiving the recommendation information transmitted by the program recommendation server, and the recommendation information reception Based on the recommendation information received in the step.
- a program control step for performing at least one of program selection, recording, playback, and erasure.
- a program included in broadcast waves or communication data is received, and input of operation information representing a user operation on the received program is accepted.
- the received operation information is communicably connected via a network, and is transmitted to a program recommendation server that creates recommendation information related to a program that meets the user's preference according to the operation information. Thereafter, the recommendation information transmitted by the program recommendation server is received, and at least one of program selection, recording, reproduction, and deletion is determined based on the received recommendation information.
- the program recommendation server is connected to be communicable with each other via a network, and represents an operation representing a user operation on the program transmitted by the program viewing terminal for viewing the received program.
- An operation information receiving unit for receiving information
- an operation information storing unit for storing the operation information received by the operation information receiving unit for each user, and the operation information received by the operation information receiving unit.
- the first grouping unit for grouping the users into a first group and the first grouping unit are grouped by the first grouping unit.
- a recommendation information generation unit that generates the recommendation information specific to the first group, and a recommendation information transmission unit that transmits the recommendation information generated by the recommendation information generation unit to the program viewing terminal.
- the program recommendation program is connected to each other via a network so that they can communicate with each other, and the number transmitted by the program viewing terminal for viewing the received program.
- An operation information receiving unit that receives operation information representing a user's operation on the set, an operation information storage unit that stores the operation information received by the operation information receiving unit for each user, and received by the operation information receiving unit
- a first grouping unit for grouping the users into a first group based on the operated operation information and the operation information for each user stored in the operation information storage unit; and the first grouping.
- a second grouping unit that groups the users into a second group that includes the first group grouped by a unit, and a user that belongs to the first group grouped by the first grouping unit And the second group grouped by the second grouping unit.
- a recommendation information generation unit that generates the recommendation information that is specific to the first group based on a result of the operation information of users belonging to the group, and the recommendation information that is generated by the recommendation information generation unit.
- a computer is caused to function as a recommendation information transmission unit that transmits to the program viewing terminal.
- a program recommendation method receives operation information representing a user operation on the program transmitted by a program viewing terminal that is connected to be communicable with each other via a network and views the received program.
- the users are grouped into the first group and include the second loop. Users are grouped into groups. Based on the totaled result of operation information of users belonging to the first group and the totaled result of operation information of users belonging to the second group, recommendation information specific to the first group was generated and generated Recommendation information is transmitted to the program viewing terminal.
- a program operated by another user having the same preference as that of the first user is extracted based on the total result of the operation information of the users belonging to the first group, and the first group including the first group is extracted. Based on the results of the operation information of users belonging to group 2, the programs that are generally preferred are extracted, and the programs extracted based on the first group are extracted based on the second group. Since the influence of the selected program is eliminated, it is possible to eliminate programs with a high degree of preference in general, and to recommend programs that are specific to specific user preferences.
- a program operated by another user having the same preference as that of the first user is extracted based on the total result of the operation information of the users belonging to the first group, and the first group is extracted.
- the programs that are generally preferred are extracted, and the programs extracted based on the first group are changed to the second group. Since the influence of the program extracted based on this is eliminated, it is possible to eliminate the program having a high degree of preference in general, and to recommend a program specialized for the taste of a specific user.
- the program recommendation system, the program viewing terminal, the program viewing program, the program viewing method, the program recommendation server, the program recommendation program, and the program recommendation method according to the present invention generally excludes programs with high preference! / A program specialized in specific user preferences, a program viewing terminal for viewing the received program, a program viewing program and a program viewing method, Useful as a program recommendation server, a program recommendation program and a program recommendation method for creating recommended information related to a user's preference according to operation information, and a program recommendation system including a program viewing terminal and a program recommendation server .
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Human Computer Interaction (AREA)
- Computing Systems (AREA)
- Computer Graphics (AREA)
- Two-Way Televisions, Distribution Of Moving Picture Or The Like (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Description
Claims
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
EP07792282.1A EP2051509B1 (en) | 2006-08-10 | 2007-08-09 | Program recommendation system, program view terminal, program view program, program view method, program recommendation server, program recommendation program, and program recommendation method |
US12/377,089 US8296803B2 (en) | 2006-08-10 | 2007-08-09 | Program recommendation system, program view terminal, program view program, program view method, program recommendation server, program recommendation program, and program recommendation method |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2006-219017 | 2006-08-10 | ||
JP2006219017 | 2006-08-10 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2008018550A1 true WO2008018550A1 (fr) | 2008-02-14 |
Family
ID=39033079
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2007/065632 WO2008018550A1 (fr) | 2006-08-10 | 2007-08-09 | Système de recommandation de programme, terminal de consultation de programme, programme de consultation de programme, procédé de consultation de programme, serveur de recommandation de programme, programme de recommandation de programme, et procédé de recommandation de programme |
Country Status (4)
Country | Link |
---|---|
US (1) | US8296803B2 (ja) |
EP (1) | EP2051509B1 (ja) |
JP (1) | JP5236700B2 (ja) |
WO (1) | WO2008018550A1 (ja) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP2197203A2 (en) | 2008-12-10 | 2010-06-16 | Samsung Electronics Co., Ltd. | Broadcast receiving apparatus and control method thereof |
JP2013508878A (ja) * | 2009-10-27 | 2013-03-07 | テレフオンアクチーボラゲット エル エム エリクソン(パブル) | 共起セレンディピティレコメンダ |
JP2020064662A (ja) * | 2019-12-27 | 2020-04-23 | Kddi株式会社 | 番組推薦装置、番組推薦方法及び番組推薦プログラム |
Families Citing this family (40)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US7559017B2 (en) | 2006-12-22 | 2009-07-07 | Google Inc. | Annotation framework for video |
US9071729B2 (en) | 2007-01-09 | 2015-06-30 | Cox Communications, Inc. | Providing user communication |
US9135334B2 (en) | 2007-01-23 | 2015-09-15 | Cox Communications, Inc. | Providing a social network |
US20090063994A1 (en) * | 2007-01-23 | 2009-03-05 | Cox Communications, Inc. | Providing a Content Mark |
JP2009043156A (ja) * | 2007-08-10 | 2009-02-26 | Toshiba Corp | 番組検索装置および番組検索方法 |
US8181197B2 (en) * | 2008-02-06 | 2012-05-15 | Google Inc. | System and method for voting on popular video intervals |
US8112702B2 (en) * | 2008-02-19 | 2012-02-07 | Google Inc. | Annotating video intervals |
US8566353B2 (en) * | 2008-06-03 | 2013-10-22 | Google Inc. | Web-based system for collaborative generation of interactive videos |
US20100011020A1 (en) * | 2008-07-11 | 2010-01-14 | Motorola, Inc. | Recommender system |
US8826117B1 (en) | 2009-03-25 | 2014-09-02 | Google Inc. | Web-based system for video editing |
US8132200B1 (en) * | 2009-03-30 | 2012-03-06 | Google Inc. | Intra-video ratings |
US8707363B2 (en) * | 2009-12-22 | 2014-04-22 | Eloy Technology, Llc | Synchronization guides for group video watching |
US20110289075A1 (en) * | 2010-05-24 | 2011-11-24 | Nelson Erik T | Music Recommender |
WO2011151817A1 (en) * | 2010-05-30 | 2011-12-08 | Honeybee Tv Israel Ltd. | Methods and system for providing time based media content suggestions |
US9167302B2 (en) | 2010-08-26 | 2015-10-20 | Cox Communications, Inc. | Playlist bookmarking |
US20120159329A1 (en) * | 2010-12-16 | 2012-06-21 | Yahoo! Inc. | System for creating anchors for media content |
JP2012137991A (ja) * | 2010-12-27 | 2012-07-19 | Jvc Kenwood Corp | 操作制御装置、操作制御プログラム、及び操作制御方法 |
US9420320B2 (en) | 2011-04-01 | 2016-08-16 | The Nielsen Company (Us), Llc | Methods, apparatus and articles of manufacture to estimate local market audiences of media content |
KR20140006063A (ko) * | 2011-04-25 | 2014-01-15 | 알까뗄 루슨트 | 추천 서비스들에서 사생활 보호 |
WO2013133879A1 (en) * | 2012-03-08 | 2013-09-12 | Thomson Licensing | A method of recommending items to a group of users |
US20130283330A1 (en) * | 2012-04-18 | 2013-10-24 | Harris Corporation | Architecture and system for group video distribution |
JP5420023B2 (ja) * | 2012-06-08 | 2014-02-19 | 株式会社Tbsテレビ | 映像関連情報送信システム及び方法、これらに用いられる関連情報送信装置 |
US8843951B1 (en) * | 2012-08-27 | 2014-09-23 | Google Inc. | User behavior indicator |
KR20140094125A (ko) * | 2013-01-21 | 2014-07-30 | 삼성전자주식회사 | 디스플레이 장치 및 이를 이용한 사용자 맞춤형 정보 제공방법 |
GB2510424A (en) * | 2013-02-05 | 2014-08-06 | British Broadcasting Corp | Processing audio-video (AV) metadata relating to general and individual user parameters |
US9319724B2 (en) * | 2013-02-28 | 2016-04-19 | Verizon and Redbox Digital Entertainment Services, LLC | Favorite media program scenes systems and methods |
JPWO2014136273A1 (ja) * | 2013-03-08 | 2017-02-09 | 株式会社東芝 | 情報配信システムおよび情報配信方法 |
WO2014142301A1 (ja) * | 2013-03-15 | 2014-09-18 | 日本電気株式会社 | 情報提供システム、サーバ装置、動画表示装置、情報提供方法、およびプログラム |
US9326026B2 (en) | 2013-10-31 | 2016-04-26 | At&T Intellectual Property I, Lp | Method and apparatus for content distribution over a network |
JP5920848B2 (ja) * | 2014-02-14 | 2016-05-18 | Necパーソナルコンピュータ株式会社 | 情報処理装置、情報処理方法、及びプログラム |
US9807436B2 (en) * | 2014-07-23 | 2017-10-31 | Rovi Guides, Inc. | Systems and methods for providing media asset recommendations for a group |
JP6711275B2 (ja) * | 2014-09-22 | 2020-06-17 | ソニー株式会社 | 画像表示制御装置、送信装置、および画像表示制御方法、並びにプログラム |
US10140379B2 (en) | 2014-10-27 | 2018-11-27 | Chegg, Inc. | Automated lecture deconstruction |
US10356458B2 (en) * | 2017-03-06 | 2019-07-16 | The Directv Group, Inc. | Controlling content presentation and display of program information in an electronic program guide |
US11601715B2 (en) | 2017-07-06 | 2023-03-07 | DISH Technologies L.L.C. | System and method for dynamically adjusting content playback based on viewer emotions |
JP6627042B2 (ja) * | 2017-09-28 | 2020-01-08 | 有限会社NetA | 情報処理装置、情報処理方法およびプログラム |
US10171877B1 (en) * | 2017-10-30 | 2019-01-01 | Dish Network L.L.C. | System and method for dynamically selecting supplemental content based on viewer emotions |
JP7137825B2 (ja) * | 2018-06-04 | 2022-09-15 | Jcc株式会社 | 映像情報提供システム |
US11847503B2 (en) * | 2020-01-28 | 2023-12-19 | Hewlett Packard Enterprise Development Lp | Execution of functions by clusters of computing nodes |
CN113382310B (zh) * | 2021-06-02 | 2023-01-03 | 北京百度网讯科技有限公司 | 信息推荐方法及装置、电子设备和介质 |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2005039749A (ja) * | 2003-07-18 | 2005-02-10 | Sony Corp | 情報処理装置および方法、記録媒体、並びにプログラム |
JP2005057713A (ja) * | 2003-08-07 | 2005-03-03 | Sony Corp | 情報処理装置および方法、プログラム、並びに記録媒体 |
Family Cites Families (16)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6005597A (en) * | 1997-10-27 | 1999-12-21 | Disney Enterprises, Inc. | Method and apparatus for program selection |
JP2000250944A (ja) * | 1998-12-28 | 2000-09-14 | Toshiba Corp | 情報提供方法、情報提供装置、情報受信装置、並びに情報記述方法 |
AU2001247245A1 (en) * | 2000-03-01 | 2001-09-12 | Expanse Networks, Inc. | Advertisment monitoring and feedback system |
JP2001298677A (ja) | 2000-04-14 | 2001-10-26 | Hirofumi Naito | 番組選択支援システム |
US20020174428A1 (en) * | 2001-03-28 | 2002-11-21 | Philips Electronics North America Corp. | Method and apparatus for generating recommendations for a plurality of users |
US20030084450A1 (en) | 2001-10-25 | 2003-05-01 | Thurston Nathaniel J. | Method and system for presenting personalized television program recommendation to viewers |
JP4009940B2 (ja) * | 2001-12-21 | 2007-11-21 | 日本電信電話株式会社 | 番組選択支援情報提供サービスシステムとサーバ装置および端末装置ならびに番組選択支援情報提供方法とプログラムおよび記録媒体 |
JP2003199084A (ja) | 2001-12-27 | 2003-07-11 | Seiko Epson Corp | 番組編集システム、番組情報検索システム、番組情報取得システム及び番組編集プログラム、並びに番組編集方法 |
US20060085818A1 (en) | 2002-11-15 | 2006-04-20 | Bodlaender Maarten P | Introducing new content items in a community-based recommendation system |
JP4329436B2 (ja) | 2003-07-08 | 2009-09-09 | ソニー株式会社 | コンテンツ視聴支援システム及びコンテンツ視聴支援方法、並びにコンピュータ・プログラム |
US7594245B2 (en) * | 2004-03-04 | 2009-09-22 | Sharp Laboratories Of America, Inc. | Networked video devices |
JP2005260647A (ja) | 2004-03-12 | 2005-09-22 | Matsushita Electric Ind Co Ltd | コンテンツ案内サービスシステム、コンテンツ案内装置、コンテンツ案内方法及びコンテンツ案内プログラム |
JP4586446B2 (ja) * | 2004-07-21 | 2010-11-24 | ソニー株式会社 | コンテンツ記録再生装置、コンテンツ記録再生方法及びそのプログラム |
US8943537B2 (en) * | 2004-07-21 | 2015-01-27 | Cox Communications, Inc. | Method and system for presenting personalized television program recommendation to viewers |
US8181201B2 (en) * | 2005-08-30 | 2012-05-15 | Nds Limited | Enhanced electronic program guides |
US8141114B2 (en) * | 2006-02-28 | 2012-03-20 | Microsoft Corporation | Content ratings and recommendations |
-
2007
- 2007-08-09 US US12/377,089 patent/US8296803B2/en active Active
- 2007-08-09 WO PCT/JP2007/065632 patent/WO2008018550A1/ja active Application Filing
- 2007-08-09 EP EP07792282.1A patent/EP2051509B1/en active Active
-
2010
- 2010-08-03 JP JP2010174640A patent/JP5236700B2/ja active Active
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2005039749A (ja) * | 2003-07-18 | 2005-02-10 | Sony Corp | 情報処理装置および方法、記録媒体、並びにプログラム |
JP2005057713A (ja) * | 2003-08-07 | 2005-03-03 | Sony Corp | 情報処理装置および方法、プログラム、並びに記録媒体 |
Non-Patent Citations (1)
Title |
---|
See also references of EP2051509A4 * |
Cited By (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP2197203A2 (en) | 2008-12-10 | 2010-06-16 | Samsung Electronics Co., Ltd. | Broadcast receiving apparatus and control method thereof |
EP2197203A3 (en) * | 2008-12-10 | 2010-07-14 | Samsung Electronics Co., Ltd. | Broadcast receiving apparatus and control method thereof |
US9094639B2 (en) | 2008-12-10 | 2015-07-28 | Samsung Electronics Co., Ltd. | Broadcasting receiving apparatus and control method thereof |
US9674477B2 (en) | 2008-12-10 | 2017-06-06 | Samsung Electronics Co., Ltd. | Broadcasting receiving apparatus and control method thereof |
US10021460B2 (en) | 2008-12-10 | 2018-07-10 | Samsung Electronics Co., Ltd. | Broadcasting receiving apparatus and control method thereof |
US10397660B2 (en) | 2008-12-10 | 2019-08-27 | Samsung Electronics Co., Ltd. | Broadcasting receiving apparatus and control method thereof |
JP2013508878A (ja) * | 2009-10-27 | 2013-03-07 | テレフオンアクチーボラゲット エル エム エリクソン(パブル) | 共起セレンディピティレコメンダ |
JP2020064662A (ja) * | 2019-12-27 | 2020-04-23 | Kddi株式会社 | 番組推薦装置、番組推薦方法及び番組推薦プログラム |
JP6998933B2 (ja) | 2019-12-27 | 2022-01-18 | Kddi株式会社 | 番組推薦装置、番組推薦方法及び番組推薦プログラム |
Also Published As
Publication number | Publication date |
---|---|
EP2051509A4 (en) | 2011-04-13 |
EP2051509A1 (en) | 2009-04-22 |
JP5236700B2 (ja) | 2013-07-17 |
US20100169927A1 (en) | 2010-07-01 |
US8296803B2 (en) | 2012-10-23 |
JP2010259105A (ja) | 2010-11-11 |
EP2051509B1 (en) | 2016-07-27 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
JP5236700B2 (ja) | 番組視聴端末、番組視聴プログラム、番組視聴方法及び番組推薦サーバ | |
JP4593603B2 (ja) | 番組推薦システム、番組視聴端末、番組視聴プログラム、番組視聴方法、番組推薦サーバ、番組推薦プログラム及び番組推薦方法 | |
US9924228B2 (en) | System and method for providing program recommendations through multimedia searching based on established viewer preferences | |
JP4783283B2 (ja) | 番組選択システム | |
US8713606B2 (en) | Systems and methods for generating a user profile based customized media guide with user-generated content and non-user-generated content | |
US9143830B2 (en) | Information processing apparatus, information processing method, computer program, and information sharing system | |
TWI306228B (en) | Information processor, informatino processing method, and recording media | |
US8793729B2 (en) | Preferential program guide | |
US8453169B2 (en) | Video output device and video output method | |
JP4458832B2 (ja) | 番組抽出装置およびその制御方法 | |
US20030237093A1 (en) | Electronic program guide systems and methods for handling multiple users | |
US9396761B2 (en) | Methods and systems for generating automatic replays in a media asset | |
KR20160132809A (ko) | 컨텐츠의 재생 제어 방법 및 이를 수행하기 위한 컨텐츠 재생 장치 | |
WO2005071951A1 (ja) | 番組推薦装置、被番組推薦装置、および番組推薦システム | |
US20140114919A1 (en) | Systems and methods for providing synchronized media content | |
KR20100069395A (ko) | 얼굴 인식 기술을 이용하여 사용자에 따라 특화된 아이피티브이 콘텐츠를 추천하는 시스템 및 방법 | |
US11974014B2 (en) | Systems and methods for dynamically educating users on sports terminology | |
US20130177288A1 (en) | Electronic device and audio output method | |
WO2011118249A1 (ja) | コンテンツ推薦サーバ、コンテンツ表示端末、およびコンテンツ推薦システム | |
KR20180049639A (ko) | 전자 장치 및 이의 제어 방법 | |
EP2884758A1 (en) | Electronic apparatus, scene list processing method and information storage medium | |
JP5198643B1 (ja) | 映像解析情報アップロード装置及び映像視聴システム及び方法 |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 07792282 Country of ref document: EP Kind code of ref document: A1 |
|
WWE | Wipo information: entry into national phase |
Ref document number: 12377089 Country of ref document: US |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
REEP | Request for entry into the european phase |
Ref document number: 2007792282 Country of ref document: EP |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2007792282 Country of ref document: EP |
|
NENP | Non-entry into the national phase |
Ref country code: RU |