JP2006127145A - Content recommendation apparatus, content recommendation method, content recommendation program, and computer-readable recording medium with program recorded thereon - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To recommend a program to a viewer by appropriately reflecting preference of the viewer in viewing the program, and by reflecting the preference varying with changes in the program content. <P>SOLUTION: This apparatus is provided with a viewer information accumulation part 105 for accumulating viewer information including inputted information about individual viewers, a content accumulation part 106 for accumulating contents and content information as information about the contents, a content reproduction processing part 113 for reproducing one or more contents accumulated in the content accumulation part 106, a preference degree calculation part 109 for performing calculation of the preference degree for a plurality of times by using the viewer information and the content information of the identical contents being reproduced by the content reproduction processing part 113, and a recommended content deciding part 111 for deciding contents to be recommended to the viewers by referring to the calculated preference degree every time the preference degree for the contents being reproduced is calculated by the preference degree calculation part 109. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to a content recommendation device, a content recommendation method, a content recommendation program, and a computer-readable recording medium on which the content is recommended.

  Currently, devices that search for and recommend TV programs and Internet broadcast programs that match the viewer's preference and automatically record them are known. These search for information that matches the viewer's preference information from the program information given to each program.

  For example, Patent Literature 1 calculates the viewer's preference for each program, determines the program with the highest preference among the currently broadcast programs, and presents the program name to the viewer. An automatic program video selection device that automatically starts playback of the program is disclosed.

Further, in Patent Document 2, the viewer's preference for each program is not calculated for each program, but the viewer's preference for the program is calculated at a predetermined time, and the calculated preference is added. A program information providing apparatus that transmits the program (content) to a viewer-side terminal is disclosed. In the case of this patent document 2, the terminal on the viewer side checks the preference level transmitted from the program information providing apparatus, determines the program with the highest preference level at present, and presents the program name to the viewer. Or automatically start playing the program.
JP 2003-289481 A (released on October 10, 2003) JP 2004-15529 A (published on January 15, 2004)

  However, in Patent Document 1, the viewer's preference for each program is calculated for each program, but the viewer's preference for these programs is not simply given uniformly for each program. That is, the degree of preference of the viewer for the program changes as the program content changes during the broadcast.

  For example, in a baseball program, the viewer's preference for the program differs depending on whether the team the viewer is supporting is winning or losing. In addition, the degree of preference of the viewer for the program differs depending on whether it is a perfect chance for the team that the viewer is supporting or a pinch of desperate life.

  Furthermore, in a song program, the viewer's preference for the program varies depending on the singer currently singing.

  On the other hand, in Patent Document 2, the viewer's preference for each program is not calculated for each program, but is calculated for each predetermined time, so that the program progresses as compared with the configuration of Patent Document 1. It can be said that it is preferable in that the degree of preference of the viewer for the program can be calculated by reflecting the change of the program content. However, the preference level is calculated by the program providing device, and this program providing device adds the calculated preference level and transmits the program. For this reason, the program providing apparatus needs to change the transmission content for each viewer, and the burden is large.

  Furthermore, when a program is recorded and the program is viewed at a later date, the preference level at the time of transmission from the program providing device is given to the recorded program. It is difficult to appropriately reflect the degree of preference when watching the program. That is, when the transmitted program is recorded and the program is viewed at a later date, the preference level when viewing the program is different from the preference level when transmitting the program. For example, in a highly current program such as a news program, even if the preference at the time of program transmission is high, the preference may decrease according to the number of days that have elapsed since the program was transmitted.

  SUMMARY OF THE INVENTION Therefore, an object of the present invention is to provide a content recommendation device and content that can appropriately reflect the degree of preference of the viewer at the time of viewing the program and can recommend a viewing program that reflects the degree of preference that changes as the program content changes. It is an object of the present invention to provide a recommendation method, a content recommendation program, and a computer-readable recording medium in which the recommendation method is recorded.

  The content recommendation device of the present invention includes a viewer information storage unit that stores viewer information including information about individual viewers, a content storage unit that stores content and content information as information about the content, and content storage The content playback means for playing back one or more contents stored in the means and the same content being played back by the content playback means using the viewer information and the content information of the content, The preference level calculating means for calculating the number of times, and each time the preference level for the content currently being played is calculated by the preference level calculation means, the calculated preference level is referred to and the content recommended to the viewer is determined. Recommended content determination means.

  The content recommendation method of the present invention includes a viewer information accumulation step for accumulating viewer information including information relating to individual viewers, a content accumulation step for accumulating content and content information as information relating to the content, and a content accumulation step. A content playback step for playing back one or more contents stored by the user, and for the same content played back by the content playback step, using viewer information and content information of the content, A preference level calculation step for performing calculation multiple times, and a recommendation for referring to the calculated preference level and determining content to be recommended to the viewer each time the preference level for the content currently being played is calculated by the preference level calculation step. A content determination step.

  According to the above configuration, one or more contents are reproduced, and for the content being reproduced, the same information is obtained using the viewer information regarding the viewer who views the content and the content information regarding the content. Then, the degree of preference is calculated a plurality of times, and the content recommended to the viewer is determined with reference to the calculated degree of preference each time.

  Here, the content information related to the content may include, for example, the content information of the content such as the title of the content, performer information, and keywords, the broadcast start time of the content, the broadcast end time of the content, and the like. .

  The viewer information may include, for example, viewer attribute information, viewer preference information, and the like. The viewer attribute information may include information such as the viewer's name, the viewer's address, the viewer's age, and the viewer's sex. The viewer attribute information may be input by the viewer himself / herself, or personal information managed elsewhere, such as personal information management software, may be used.

  As described above, the preference level is calculated multiple times for the same content being played, and the content is recommended to the viewer each time, thereby reflecting the viewer's preference that changes as the content changes. Thus, it is possible to accurately determine and recommend content that suits the viewer's preference.

  Note that the content recommendation device may include viewer identification means for identifying the viewer based on the viewer identification information input from the viewer. In this case, by identifying the viewer, the information of the viewer information storage unit and the content information storage unit is managed for each viewer. Accordingly, even when the content recommendation device is used by a plurality of people, by identifying the viewer who currently uses the content recommendation device, the preference degree of the content is calculated for each viewer, and for each viewer Can recommend content.

  In this case, in the program information providing device disclosed in Patent Document 2, content is recommended based on the preference of the viewer at the time of content transmission, whereas in the content recommendation device of the present invention, the viewer views the content. Since the content is recommended based on the preference level of the viewer at the time of viewing, the viewer's preference at the time of viewing the content can be appropriately reflected.

  The content recommendation apparatus preferably includes output means for outputting recommended content for the viewer determined by the recommended content determination means.

  According to the above configuration, the viewer can easily view the content that is output from the output means and matches the user's preference.

  More preferably, the content recommendation device further includes output means for outputting information that can identify the recommended content for the viewer determined by the recommended content determination means.

  According to the above configuration, the viewer can select the content that matches the user's preference based on the information that can be used to identify the recommended content, for example, the information indicating the recommended content name or the content of the recommended content. Can be watched easily.

  In the above content recommendation device, preferably, the output unit further outputs information indicating a reason for recommending the recommended content to the viewer determined by the recommended content determination unit.

  According to the above configuration, the viewer can know why the content is recommended based on the information output from the output means and indicating the reason for recommending the content, and understands the reason for recommendation. You can watch the recommended content above.

  In the above content recommendation device, preferably, the output unit further outputs information indicating the degree of preference of the recommended content to the viewer determined by the recommended content determination unit.

  According to the above configuration, the viewer can know his / her preference degree with respect to the recommended content calculated by the content recommendation device, and can view the recommended content after knowing this preference degree.

  Preferably, the content recommendation device includes an output unit that outputs information indicating the preference level calculated by the preference level calculation unit for the content when a plurality of content items are played back by the content playback unit. I have.

  According to said structure, an output means outputs the information which shows the preference calculated by the preference calculation means. Thereby, the viewer can know the degree of preference of a plurality of contents being reproduced in the content recommendation device.

  In the above content recommendation device, preferably, the output unit outputs information indicating a transition of the preference level calculated by the preference level calculation unit for the recommended content.

  According to said structure, the viewer can know the temporal change of the preference degree with respect to a recommendation content.

  In the above content recommendation device, preferably, the output unit is information indicating a transition of the preference level calculated by the preference level calculation unit for the content when a plurality of contents are played back by the content playback unit. Is output.

  According to the above configuration, the information indicating the transition of the preference level for the plurality of contents played back by the content playback unit is output from the output unit, so that the viewer can determine the preference level for the plurality of content being played back. It is possible to know temporal changes and also to know the degree of preference of each content at a certain time.

  In the above content recommendation device, preferably, the preference level calculating means calculates the preference level for the same content every predetermined time, and the recommended content determining means is the content recommended to the viewer every predetermined time. To decide.

  According to the above configuration, the preference level calculation unit repeatedly calculates the preference level of the same content every predetermined time, for example, every 5 minutes or every 10 minutes, and the recommended content determination unit recommends to the viewer each time. Decide what content to use. Therefore, it is possible to accurately recommend the content that suits the viewer's preference not in units of content but in smaller units of predetermined time. In other words, the viewer's preference level for content changes according to changes in the content content as time progresses, so the preference content of the same content is repeatedly calculated every predetermined time to determine the recommended content. Thus, even if the content changes with time, it is possible to accurately recommend content that matches the viewer's preference.

  Note that calculating the degree of preference every predetermined time may include any case of calculating the degree of preference within a predetermined time every predetermined time and calculating the degree of preference at that time every predetermined time. .

  In the above content recommendation device, preferably, the content is composed of a plurality of parts, the preference level calculating means calculates the preference level for each part of the same content, and the recommended content determining means is the viewing for each part. Content to recommend to the user.

  Note that, for example, partial information is given to each part constituting the content, and by using the partial information, the preference level calculation unit is configured to calculate the preference level of the content for each part. May be. The partial information may be information indicating, for example, the performer of the part, the content of the part, the keyword of the part, the genre of the part, and the like.

  According to the above configuration, the recommended content for the viewer is determined for each of a plurality of portions constituting the content more finely, not in units of content. Therefore, even if the degree of preference of the viewer for the content changes according to the change in the content of each part constituting the content, the content that matches the viewer's preference can be recommended accurately.

  In the above content recommendation device, preferably, the content stored by the content storage means is content currently being broadcast, content broadcast in the past, content read from an external recording medium, or downloaded via a network It is at least one type of content.

  As described above, the content stored by the content information storage unit is not limited to the content currently being broadcast, and may be content broadcast in the past or content read from an external recording medium. Alternatively, it may be content distributed and downloaded via a network such as video on demand (VoD). However, the network may be wired or wireless. The content refers to all multimedia contents including not only video media recorded on television programs and recording media but also music media such as radio programs and music.

  In the content recommendation device, preferably, the viewer preference information included in the viewer information includes the viewer's favorite genre, the viewer's favorite person, the viewer's favorite keyword, and the viewer's preference. At least one of the content preferred by the viewer and the viewing history information of the past content of the viewer.

  Thus, the content recommendation device recommends content suitable for the viewer by registering at least one information of the favorite genre, favorite person, favorite keyword, and favorite content by the viewer. be able to. Further, by using the viewing history information of the viewer's past content, the preference level can be calculated more accurately, and the content preferable for the viewer can be recommended.

  Note that the preference information includes, for example, programs that the viewer does not want to see, keywords that the viewer does not want to see, genres that the viewer does not want to see, and performers that the viewer does not want to see. Such information may also be included. Furthermore, viewer preference information determined from action history information such as operation history information by viewers during content viewing and viewer behavior during content recording may also be included. Also, the preference information may be input by the viewer himself, or the viewer preference information may be estimated from other information so as to determine the viewer preference information from the operation history information and the action history information. Absent.

  The content recommendation device preferably includes a receiving unit that receives content, and a comparison unit that compares a preference level calculated by the preference level calculation unit with a predetermined threshold, and the recommended content determination unit is a comparison unit. Thus, when it is determined that the preference level of the content that can be currently received by the receiving means does not exceed a predetermined threshold, the first content that can be received by the receiving means among the contents that are being played by the content playing means The content with the highest preference is recommended from the second content other than.

  According to the above configuration, if the preference level of the currently received content (first content) does not satisfy the predetermined threshold in the receiving unit, the content other than the first content in the currently played content The content with the highest preference is recommended from the second content that is the content.

  As a result, even if all the contents currently being received by the receiving means do not meet the viewer's preference, it is possible to recommend the content that matches the viewer's preference from the currently reproduced content.

  In the above content recommendation device, preferably, when the recommended content is determined from the second content, the recommended content is determined from the content whose past number of times of viewing is a predetermined threshold value or less.

  Thereby, when determining the recommended content from the second content, it is possible to determine whether or not the content is to be the recommended content in consideration of the number of times the viewer has viewed the content in the past. For this reason, for example, a threshold value can be set, and for example, only content whose past number of times of viewing is less than one can be targeted for recommended content.

  The content recommendation apparatus preferably includes a comparison unit that compares the preference level calculated by the preference level calculation unit with a predetermined threshold value, and the recommended content determination unit selects content whose preference level does not exceed the threshold value as recommended content. Exclude from candidates.

  According to the above configuration, the comparison unit compares the degree of preference of the content with a predetermined threshold value, and based on the result, the recommended content determination unit excludes content whose preference level does not exceed the threshold value from the recommended content candidates. To do.

  Thus, by comparing the content preference level with a predetermined threshold, it is possible to limit the recommended content to the viewer, and to prevent a situation where content with a low preference level is recommended.

  In the above content recommendation device, preferably, the comparison unit changes the threshold according to the viewer.

  According to the above configuration, the predetermined threshold is not uniquely determined by the content recommendation device, but is changed according to the viewer. Therefore, the degree of preference is reflected in the recommended content, which is different for each viewer. Thus, the content can be recommended to the viewer.

  In the above content recommendation device, the comparison unit preferably changes the threshold according to a time zone when the viewer appreciates the content.

  As a result, the threshold value can be changed according to the time period when the viewer views the content. Therefore, the threshold value should be set as the morning threshold value, the daytime threshold value, the evening threshold value, the nighttime threshold value, the midnight threshold value, etc. The threshold value can be freely selected according to the time zone. As a result, it is possible to appropriately determine the recommended content even in a time zone in which content with a high degree of preference is concentrated and vice versa.

  In the above content recommendation device, preferably, a comparison unit that compares the preference level calculated by the preference level calculation unit with a predetermined threshold value, and a display control unit that displays information output from the output unit on the display unit. The display control means includes a plurality of screens of the display means when the comparison means determines that the preference level of the plurality of contents exceeds a predetermined threshold among the plurality of contents reproduced by the content reproduction means. A plurality of contents having a preference level exceeding a predetermined threshold are distributed and displayed in these divided areas.

  According to said structure, when the preference level of several content exceeds a predetermined threshold value, the display screen of a display control means is divided | segmented into a some area | region, and the plurality of preference levels exceed a predetermined threshold value in these divided areas. Are sorted and displayed.

  Thereby, for example, even when there are a plurality of contents having a high degree of preference, all of the contents can be displayed on the screen. In addition, the size of the area to be divided at that time is not limited to equal division, and the size of the area to be divided may be changed according to the degree of preference for the content, and the content having a higher degree of preference is displayed in a larger area. You can also.

  The content recommendation device preferably includes display means for displaying information output by the output means.

  According to the above configuration, the recommended content, the information that can identify the recommended content on the display means, the reason for recommending the recommended content to the viewer, the information indicating the content preference level, or the content preference level Information indicating the transition is displayed, and the viewer can view the information.

  As described above, the content recommendation device according to the present invention has the viewer information storage unit, the content storage unit, the content playback unit, and the content information for the same content being played back by the viewer information and the content. The preference level calculation means for calculating the preference level a plurality of times using the content information, and the preference level is referred to each time the preference level for the content currently being played is calculated by the preference level calculation means. And a recommended content determining means for determining content recommended to the viewer.

  Accordingly, it is possible to accurately determine and recommend the content that matches the viewer's preference, reflecting the viewer's preference that changes as the content changes.

  Hereinafter, an embodiment of the present invention will be described with reference to the drawings. In the drawings used for the following description, the same parts are denoted by the same reference numerals. Their names and functions are also the same. Therefore, detailed description thereof will not be repeated.

(Functional configuration)
FIG. 1 shows a block diagram of a functional configuration of content recommendation apparatus 100 according to the present embodiment.

  As shown in FIG. 1, the content recommendation device 100 includes an input unit 101, a reception unit 102, an input unit 103, and a viewer recognition unit 104. Furthermore, the content recommendation device 100 includes a viewer information storage unit 105, a content storage unit 106, a viewer information extraction unit 107, a content information extraction unit 108, a preference level calculation unit 109, a preference level comparison unit 110, and a recommended content determination unit 111. A content extraction unit 112, a content reproduction processing unit 113, an output processing unit 114, a content display processing unit 115, and an output number counting unit 116.

  The content includes not only video programs recorded on television programs and recording media broadcasted via wireless or wired communication, but also music media such as radio programs and music provided in the same manner. That is, it refers to all multimedia contents provided by various forms and various media.

  The input unit 101 includes, for example, an interface that receives input from a user or an interface that receives input from an external device, and includes viewer attributes such as viewer ID, name, gender, age, and address, and viewer preference information. The viewer information 11 consisting of is input. This viewer information 11 can be input by, for example, the user himself / herself or can be input from an external device.

  Here, the viewer preference information is information such as the viewer's favorite genre, the viewer's favorite person, the viewer's favorite keyword, or the viewer's favorite content, but is not necessarily limited thereto. It is not something. For example, the viewing history information of the viewer's past contents may be included in the preference information. Furthermore, the viewer preference information includes information such as a genre that the viewer does not want to see, a person that the viewer does not want to see, a keyword that the viewer does not want to see, or content that the viewer does not want to see. It does not matter. Further, it may include viewer preference information determined from operation history information by the viewer during content viewing and behavior history information such as viewer behavior during content recording.

  The receiving unit 102 receives, for example, a television program broadcast via wireless or wired, and inputs content and content information 12. The content information is information related to the content given to the content, and includes, for example, the content information of the content, such as the title of the content, performer information, and keywords, the broadcast start time of the content, the broadcast end time of the content, etc. It does not matter. These content and content information 12 can be obtained from content broadcast by a content provider via wireless or wired communication. The content information 12 is, for example, EPG (Electric Program Guide: electronic program guide).

  The input unit 103 is for inputting, for example, a television program input via a recording medium, and inputs content and content information. These contents and content information can be obtained from the contents stored in the recording medium in the external device.

  The viewer recognition unit 104 receives the viewer identification information 14 for identifying the viewer, and recognizes the viewer. The viewer identification information 14 is, for example, a viewer name or a viewer identification number, and can be obtained by the user inputting or selecting such information.

  Note that the viewer identification information 14 and the viewer identification method are not limited to these. For example, human body information such as a viewer's fingerprint and iris is registered in advance, and the information is input when the content recommendation device 100 is used, and compared with information registered in advance in the viewer recognition unit 104. Thus, a method for identifying the viewer may be used.

  The viewer information accumulation unit 105 manages the viewer information 11 input via the input unit 101.

  When the viewer is recognized by the viewer recognition unit 104, the viewer information storage unit 105 manages the viewer information for each recognized viewer. In this way, by managing each viewer, the content recommendation device 100 can be used by a plurality of people.

  The content storage unit 106 stores and manages content and content information 12 input via the receiving unit 102 or the input unit 103.

  When a viewer is recognized by the viewer recognition unit 104, the content storage unit 106 manages content and content information 12 for each recognized viewer. Thus, by managing the content and the content information 12 for each viewer, even when the content recommendation device 100 is used by a plurality of people, the viewer who is currently using the content recommendation device 100 is recognized. The content input by the recognized viewer and the content information 12 can be extracted. This makes it possible to provide a preferred content recommendation device that can be used by multiple people.

  The viewer information extraction unit 107 extracts viewer information 11 such as viewer attributes and viewer preference information stored in the viewer information storage unit 105. When the viewer is recognized by the viewer recognition unit 104, the viewer information 11 related to the recognized viewer is extracted.

  The content information extraction unit 108 extracts content information attached to the content stored in the content storage unit 106. When the viewer is recognized by the viewer recognition unit 104, the content information attached to the content stored in the content storage unit 106 by the recognized viewer is extracted.

  The preference level calculation unit 109 uses the information extracted by the viewer information extraction unit 107 and the content information extraction unit 108 to calculate the preference level of the viewer repeatedly for the same content, and recommends it. It is passed to the content determination unit 111.

  The preference level comparison unit 110 compares the content preference level calculated by the preference level calculation unit 109 with a predetermined threshold value and passes the comparison result to the recommended content determination unit 111.

  Note that the predetermined threshold is not limited to a case where the predetermined threshold is uniquely determined in the content recommendation device, and may be changed according to the viewer, or may be changed according to the time zone when the viewer appreciates the content. It doesn't matter.

  The recommended content determination unit 111 determines the content recommended for the viewer based on the preference level calculated by the preference level calculation unit 109 and the comparison result in the preference level comparison unit 110. Note that the recommended content determining unit 111 does not necessarily have one content. A plurality of contents may be recommended to the viewer.

  The content extraction unit 112 extracts the content stored in the content storage unit 106. In this case, the content to be extracted may be selected by the viewer or may be selected by the content recommendation device 100.

  The content reproduction processing unit 113 performs reproduction processing of one or more contents extracted by the content extraction unit 112.

  The output processing unit 114 outputs the content reproduced by the content reproduction processing unit 113 to the external device 15. Furthermore, the output processing unit 114 also outputs the preference level calculated by the preference level calculation unit 109 and the reason for determining the recommended content by the recommended content determination unit 111 to the external device 15.

  The content display processing unit 115 displays information output to the external device 15 by the output processing unit 114 on the display device 16.

  The output count unit 116 counts the number of times the output processing unit 114 outputs to the external device 15 or the content display processing unit 115 displays the number of times displayed on the display device 16, and the content storage unit 106 displays the count result as the content. Are managed in association with each other.

(Configuration of a computer system for realizing the content recommendation device 100)
The content recommendation device 100 according to the present embodiment is substantially realized by computer hardware, a program executed by the computer hardware, and data stored in the computer hardware. FIG. 2 shows an internal configuration of the computer system 200.

  Referring to FIG. 2, the computer system 200 includes a computer 220 having an optical disk drive 201 and a magnetic disk drive 202, a monitor 203, and a remote controller (hereinafter referred to as “remote controller”) 204.

  In addition to the optical disk drive 201 and the magnetic disk drive 202, the computer 220 includes a remote control interface (I / F) 205 that receives a signal from the remote control 204, a CPU (central processing unit) 206, a remote control interface 205, a CPU 206, and an optical disk. A bus 207 connected to the drive 201 and the magnetic disk drive 202, a read-only memory (ROM) 208 connected to the bus 207 and storing a boot-up program and the like, and also connected to the bus 207, program instructions, system programs, And a random access memory (RAM) 209 for storing work data and the like.

  Although not shown here, the computer 220 may further include a network adapter board that provides a connection to a local area network (LAN).

  A program for causing the computer system 200 to realize the function of the content recommendation device 100, the viewer information 11 (see FIG. 1), the content and content information 12, and the viewer identification information 14 are all the optical disc drive 201 or the magnetic disc drive. The data is stored in the optical disk 210 or the magnetic disk 211 inserted in 202, and further transferred to the hard disk 212. Alternatively, the program, the viewer information 11 (see FIG. 1), the content and content information 12, and the viewer identification information 14 may be transmitted to the computer 220 through a network (not shown) and stored in the hard disk 212. The program is loaded into the RAM 209 when executed. The program may be loaded directly into the RAM 209 from the optical disk 210, from the magnetic disk 211, or via a network.

  This program includes a plurality of instructions for causing the computer 220 to realize the functions of the content recommendation device 100 according to the present embodiment. Some of the basic functions necessary to realize these functions are provided by an operating system (OS) or a third-party program running on the computer 220 or various toolkit modules installed on the computer 220. Therefore, this program does not necessarily include all functions necessary for realizing the functions of the content recommendation device 100 according to this embodiment. This program may include only instructions for executing control of the content recommendation device 100 by calling an appropriate function or “tool” in a controlled manner so as to obtain a desired result. That's fine. The operation of computer system 200 is well known and will not be repeated here.

  Note that a recording medium on which the program for realizing the functions of the content recommendation device 100, the viewer information 11 (see FIG. 1), the content and the content information 12, or the viewer identification information 14 is recorded is a CD-ROM (compact). It is not limited to the optical disk 210 such as a disk read-only memory), MO (Magneto-Optical disc), MD (Mini Disc), and DVD (Digital Versatile Disc), and a magnetic disk 211 such as an FD (flexible disk) and a hard disk. Tapes such as magnetic tapes and cassette tapes, card-type recording media such as IC (Integrated Circuit) cards and optical cards, and semiconductors such as mask ROM, EPROM (Erasable Programmable ROM), EEPROM (Electrically Erasable Programmable ROM), and flash ROM It can be either memory. However, the computer system 200 needs to be equipped with a reading device for reading from these recording media.

(Outline of main operations)
FIG. 3 is a flowchart showing an outline of main operations of the content recommendation device 100.

  Referring to FIG. 3, when the content recommendation device 100 is turned on, the program 300 is activated and the process proceeds to step S301.

  In step S301, the viewer information extraction unit 107 determines whether or not viewer information exists. When the viewer information exists, the process proceeds to step S302, and when it does not exist, the process proceeds to step S303.

  In step S302, the viewer information extraction unit 107 extracts viewer information from the viewer information storage unit 105, and the process proceeds to step S303.

  In step S303, the preference level calculation unit 109 determines whether or not the preference levels of all contents have been calculated. If the preference levels of all the contents have been calculated, the process proceeds to step S305. If the preference levels of all the contents have not been calculated, the process proceeds to step S304.

  In step S304, the preference level calculation unit 109 calculates the preference level of the content, and returns the process to step S303. Details of the preference level calculation process will be described later.

  In step S305, the recommended content determination unit 111 uses the content preference calculated in step S304 to determine content recommended for the viewer, and the process proceeds to step S306. Details of the recommended content determination process will be described later.

  In step S306, whether the output processing unit 114 performs output processing to the external device 15, or whether the content display processing unit 115 performs display processing on the display device 16 of the information output by the output processing unit 114. Judgment is made. If the output process or the display process is to be performed, the process proceeds to step S307. If not, the process ends.

  In step S307, the output processing unit 114 or the content display processing unit 115 performs output / display processing on the external device 15 or the display device 16, and ends the processing. Details of the output / display process will be described later.

(Preference level calculation processing)
FIG. 4 is a flowchart showing details of the preference level calculation process executed in step S304 shown in FIG.

  Referring to FIG. 4, when preference level calculation processing 350 is started, in step S351, content information extraction unit 108 extracts content information related to the content from content storage unit 106, and the process proceeds to step S352.

  In step S352, the preference level calculation unit 109 determines the timing for calculating the preference level of the content. When the preference level is calculated every predetermined time such as every 5 minutes or every 10 minutes, the process proceeds to step S353, and when the preference level is calculated for each part, the process proceeds to step S354. For example, as shown in FIG. 23 (a), when the preference level is calculated every 10 minutes in a certain 60-minute content, the process proceeds to step S353, and as shown in FIG. When the preference level is calculated for each constituent part, the process proceeds to step S354.

  In step S353, the preference level calculation unit 109 calculates the preference level for the content at predetermined time intervals using the content information read in step S351 and the viewer information read in step S302 shown in FIG. The process is terminated. In addition, although the calculation method of a preference degree is illustrated below, it is not necessarily limited to this.

  For example, if the viewer information includes information such as content name, keyword, performer, or genre that the viewer likes, the viewer information is compared with the content information of the content, By converting it, it may be calculated as a preference level. For example, as shown in FIG. 24, since there is a common “Mr. X” in the content information and the viewer information, when the preference level is increased by 10 points for one item match, the preference level is here. Becomes 10.

  In step S354, the preference level calculation unit 109 uses the content information read in step S351 and the viewer information read in step S302 shown in FIG. 3, and sets the preference level for the content in the content. Every time it is calculated, the process ends.

  Note that the preference degree calculation method is not limited to the above-described method.

(Recommended content decision processing)
FIG. 5 is a flowchart showing details of the recommended content determination process executed in step S305 shown in FIG. Referring to FIG. 5, when recommended content determination processing 400 is started, in step S401, recommended content determination unit 111 determines whether the content is currently being received. If the content is currently being received, the process proceeds to step S402; otherwise, the process proceeds to step S404.

  In step S402, the preference level comparison unit 110 determines whether the preference level of the content calculated by the preference level calculation unit 109 is greater than a predetermined threshold. If the preference level of the content is larger than the predetermined threshold value, the process proceeds to step S403, and if smaller, the process proceeds to step S404.

  In step S403, the recommended content determination unit 111 sets the content as a recommended content candidate, and proceeds to step S404.

  In step S404, the recommended content determination unit 111 determines whether or not the preference levels and threshold values of all the content items have been checked. If the process has been performed for all contents, the process proceeds to step S405. If not, the process returns to step S401.

  In step S405, the recommended content determination unit 111 determines whether there is a recommended content candidate determined in step S403. If there is a recommended content candidate, the process proceeds to step S406, and if not, the process proceeds to step S407.

  In step S406, the recommended content determination unit 111 determines recommended content from the recommended content candidates, and ends the process.

  Here, as a method of determining recommended content from recommended content candidates, for example, there is a method of determining content with the highest degree of preference as recommended content, but the method is not necessarily limited thereto.

  Furthermore, the recommended content is not limited to one and may be plural.

  In step S407, the recommended content determination unit 111 determines recommended content from content that is not currently being received, that is, content that is already stored in the content storage unit 106, and ends the process.

  Here, as a method of determining recommended content from content that is not currently being received, for example, there is a method of determining content with the highest degree of preference as recommended content, but this is not necessarily limited thereto.

  Furthermore, the recommended content is not limited to one and may be plural.

(Output / display processing)
FIG. 6 is a flowchart showing details of the output / display processing executed in step S307 shown in FIG. Referring to FIG. 6, when output / display processing 450 is started, in step S451, output processing unit 114 sets the output format of the recommended content determined in the processing of step S406 or step S407 shown in FIG. select. If the recommended content itself is output, the process proceeds to step S452, the process proceeds to step S453 when the information such as the title name of the recommended content can be specified, and the process proceeds to step S454 when the recommended content is not output.

  In step S452, the output processing unit 114 extracts the content that is being reproduced by the content reproduction processing unit 113 and the content determined by the recommended content determination unit 111, and proceeds to step S454.

  In step S453, the output processing unit 114 extracts information that is the content being played back by the content playback processing unit 113 and can identify the content determined by the recommended content determination unit 111, and proceeds to step S454. The information that can identify the recommended content is, for example, the title name of the recommended content, and the information is extracted from the content information stored in the content storage unit 106.

  In step S454, the output processing unit 114 determines whether or not to output the reason for determining the recommended content in the process of step S406 or step S407 shown in FIG. This determination may be performed based on an output request from the viewer, or may be performed based on an automatic determination function on the content recommendation device side. As a result of this determination, if the reason is output, the process proceeds to step S455, and otherwise, the process proceeds to step S456.

  Here, the information as the reason for determining the recommended content is not necessarily limited to any form. For example, it may be the most important keyword that has come to recommend the content.

  In step S455, the output processing unit 114 extracts the reason for determining the recommended content from the recommended content determination unit 111, and the process proceeds to step S456.

  In step S456, the output processing unit 114 determines whether to output the preference degree calculated in step S353 illustrated in FIG. 4 or step S354. This determination may be performed based on an output request from the viewer, or may be performed based on an automatic determination function on the content recommendation device side. As a result of the determination, if the preference level is output, the process proceeds to step S457, and if not output, the process proceeds to step S458.

  In step S457, the output processing unit 114 extracts the preference level for the content from the preference level calculation unit 109, and proceeds to step S458.

  In step S458, the output processing unit 114 determines whether or not to output the transition of the preference level of the content. This determination may be performed based on an output request from the viewer, or may be performed based on an automatic determination function on the content recommendation device side. As a result of the determination, if the preference degree transition is output, the process proceeds to step S459; otherwise, the process proceeds to step S460.

  In step S459, the output processing unit 114 extracts the transition of the preference level for the content from the preference level calculation unit 109, and the process proceeds to step S460.

  In step S460, the output processing unit 114 determines whether there is a recommended content other than the content. If no other recommended content exists, the process proceeds to step S461. If any other recommended content exists, the process returns to step S451.

  In step S461, the output processing unit 114 reads the output format, and proceeds to step S462.

  Here, the output format defines what format the output processing unit 114 outputs to the external device 15. For example, the layout of each extracted information may be specified.

  In step S462, the output processing unit 114 outputs to the external device 15 according to the output format read in step S461, and proceeds to step S463.

  In step S463, the content display processing unit 115 determines whether to output the information output from the output processing unit 114 to the display device 16. When displaying on the display apparatus 16, it progresses to step S464, and when not displaying, a process is complete | finished.

  In step S464, the content display processing unit 115 reads the display format, and proceeds to step S465.

  Here, the display format defines what format the content display processing unit 115 displays on the display device 16. For example, the content display processing unit 115 forcibly outputs and displays the content itself on the display device 16, or displays only the content name instead of forcibly outputting and displaying the content itself on the display device 16. It does not matter. Moreover, you may prescribe | regulate the layout at the time of displaying on the display apparatus 16, and the like.

  In step S465, in accordance with the display format read in step S464, the content display processing unit 115 displays on the display device 16, and the process ends.

  Hereinafter, examples 1 to 3 will be exemplified as specific operations of the content recommendation device 100.

Example 1
In the first embodiment described below, an example of an operation in the case where the content preference level is calculated for each part of the currently broadcast content and the content is recommended to the viewer A based on the calculated preference level will be described.

  In addition, the said part is one when one content is divided | segmented into the block of several information by the difference in the content, for example, and one content is comprised from one or more parts. Furthermore, each part shall hold | maintain the partial information which shows the content of the said part. Here, as the partial information held by each part, there are a genre to which the part belongs, a performer of the part, a keyword indicating the content of the part, information indicating at least one of the start and end of each part, and the like. However, it is not necessarily limited to these.

  FIG. 7 shows the preference information of the viewer A stored in the viewer information storage unit 105.

  In FIG. 7, viewer A's preference genre 501, performer 502, and keyword 503 are shown as viewer A's preference information 500. For example, as the preference information of the viewer A, “genre Y” is registered as a genre, five people including “Mr. A” as performers, and five keywords such as “K1” are registered as keywords. However, viewer preference information is not limited to these.

  FIG. 8 shows the details of the content currently broadcast for one hour from 7:00 to 8:00. Here, it is assumed that there are only three contents “content A”, “content B”, and “content C” that are currently being broadcast and can be received by the receiving means.

  In FIG. 8, the details of “content A” 521, “content B” 522, and “content C” 523 are shown according to the time flow 520. Each content is composed of a plurality of parts, and each part holds partial information indicating the contents of the part as described above. Here, the partial information held by each part is a part name 524, a genre name 525 to which the part belongs, a performer 526 of the part, and a keyword 527 related to the part. Is displayed.

  For example, “content A” is composed of five parts in the contents from “7:00” to “8:00”, and the part name “A-1” is from “7:00” to “7: 10 ”, the genre of the part name“ A-1 ”belongs to“ genre X ”, the performers of the part are“ Mr. A ”and“ Mr. B ”, and the keyword related to the part is“ K1 ” "And" K2 ". Note that these pieces of partial information are given for each portion when the content is distributed.

  The partial information held by the part is not limited to these. For example, the contents of the part may be held. Furthermore, the structure of the part is not limited. A part may be configured every 10 minutes like “content B” 522, or a part may be configured every 30 minutes like “content C” 523. Furthermore, a configuration in which a 10-minute portion or a 20-minute portion is mixed as in “content A” 521 may be used.

  Here, when the viewer A views the content from “7:00”, the operation when the content recommendation device 100 calculates the preference for each part of the content will be described according to the flowchart shown in FIG.

  In step S <b> 541, the viewer information extraction unit 107 determines whether or not the viewer information related to the viewer A exists in the viewer information storage unit 105.

  Since the viewer A's preference information 500 shown in FIG. 7 is managed by the viewer information storage unit 105, the process proceeds to step S542. In step S542, the viewer information extraction unit 107 reads the preference information 500 of the viewer A shown in FIG.

  In step S543, it is determined whether or not the preference levels of the three contents “content A” to “content C” have been calculated.

  In step S544, the content information extraction unit 108 reads the content information shown in FIG.

  In step S545, the preference level calculation unit 109 calculates the preference level of each content.

  Here, as a method of calculating the preference level by the preference level calculation unit 109, the preference information 500 of the viewer A is compared with each piece of partial information held by each portion. The degree shall be increased by 10. For example, in the partial name “A-1”, two items of “Person A” as the performer and “K1” as the keyword match the preference information 500 of the viewer A, so the preference level of the partial name “A-1” Is 20.

  Similarly, the processing from step S543 to step S545 is repeated, and the preference levels of all contents are calculated. The result of the calculation is shown in FIG.

  In FIG. 10, the preference 564 of each part of “content A” 561, “content B” 562, and “content C” 563 is shown according to the time flow 560. For example, the preference level of the partial name “A-1” is 20, and the preference level of the partial name “B-1” is 40.

  Note that the preference level calculation method is not limited to this.

  For example, in the preference information 500 of the viewer A shown in FIG. 7, each of the favorite genre 501, the favorite performer 502, and the favorite keyword 503 may be weighted to calculate the preference level. . For example, it is assumed that the viewer A gives priority to the favorite performer 502 among these three, and then gives priority to the favorite keyword.

  At this time, in step S545, as a method for the preference level calculation unit 109 to calculate the preference level, the preference information 500 of the viewer A is compared with each piece of partial information held by each portion, and there is one item that matches. The preference level is increased by 10, but when the favorite performer 502 matches, the weight is multiplied by 1.1, and when the favorite keyword 503 matches, the weight is multiplied by 0.8, and the favorite genre When the values match at 501, 0.5 is applied as a weight. FIG. 11 shows a calculation result of the preference degree for each part based on such a rule.

  For example, in the partial name “A-1”, “Mr. A” as a performer matches the preference information 500 of the viewer A, so 1.1 is weighted and the preference level is increased by 11. Further, since “K1” as a keyword matches the preference information 500 of the viewer A, weighting of 0.8 is performed and the preference level is increased by 8. Therefore, the preference level of the partial name “A-1” is 19.

  Similarly, in the partial name “B-4”, since “genre Y” in the genre matches the preference information 500 of the viewer A, 0.5 is weighted to increase the preference level by 5. Further, since “Mr. F” and “Mr. J” as performers coincide with the preference information 500 of the viewer A, weighting 1.1 is performed and the preference level is increased by 22. Furthermore, since “K6” and “K7” in the keyword match the preference information 500 of the viewer A, each weight is 0.8 and the preference is increased by 16. Therefore, the preference level of the partial name “B-4” is 43.

  Similarly, the processing from step S543 to step S545 is repeated, and the preference levels of all contents are calculated.

  In FIG. 11, the preference level 584 of each part of “content A” 581, “content B” 582, and “content C” 583 is shown according to the time flow 580. For example, the preference level of the partial name “A-1” is 19, and the preference level of the partial name “B-1” is 32.

  Moreover, as a method of calculating the preference level, each item of the preference information of the viewer A shown in FIG. 7 may be weighted to calculate the preference level.

  FIG. 12 is a diagram in which weighting is applied to each item of the preference information of the viewer A stored in the viewer information storage unit 105.

  In FIG. 12, viewer A's favorite genre 601, viewer A's favorite performer 602, and viewer A's favorite keyword 603 are shown as the viewer A's preference information 600. Furthermore, in each, the item 604 and the weighting 605 assigned to the item 604 are displayed in association with the item. For example, in the performer, “Mr. A” is multiplied by 1.3 as a weight, and “Mr. F” is multiplied by 0.8. However, how to give weight is not limited to this.

  In this case, as a method for the preference level calculation unit 109 to calculate the preference level in step S545, the preference information 600 of the viewer A is compared with each piece of partial information held by each portion, and if there is a matching item. Assume that the degree of preference is increased by one. Furthermore, the preference level is calculated in consideration of the weighting given to each item. FIG. 12 shows the calculation result of the preference degree for each part based on such a rule.

  For example, in the partial name “A-1”, “Mr. A” as a performer matches the preference information 600 of the viewer A, so 1.3 is weighted and the preference level is increased by 13. Further, since “K1” as a keyword matches the preference information 600 of the viewer A, 1.5 is weighted to increase the preference level by 15. Therefore, the preference level of the partial name “A-1” is 28.

  Similarly, in the partial name “B-4”, since “genre Y” in the genre matches the preference information 600 of the viewer A, 0.5 is weighted and the preference is increased by 5. Furthermore, since “Mr. F” as a performer matches the preference information 600 of the viewer A, weighting 0.8 is performed and the preference level is increased by 8. Furthermore, since “J” as a performer matches the preference information 600 of the viewer A, weighting 1.1 is performed and the preference level is increased by 11. Further, since “K6” as the keyword matches the preference information 600 of the viewer A, weighting 0.6 is performed and the preference level is increased by 6. Further, since “K7” as a keyword matches the preference information 600 of the viewer A, weighting 0.8 is performed and the preference level is increased by 8. Therefore, the preference level of the partial name “B-4” is 38.

  Similarly, the processing from step S543 to step S545 is repeated, and the preference levels of all contents are calculated.

  In FIG. 13, the degree of preference 624 of each part of “content A” 621, “content B” 622, and “content C” 623 is shown according to the time flow 620. For example, the preference level of the partial name “A-1” is 28, and the preference level of the partial name “B-1” is 27.

  Here, as described above, the three methods of calculating the preference are shown, but the method is not necessarily limited to these.

  Next, the operation of the content recommendation device 100 when determining the content recommended for the viewer A based on the degree of preference for each part shown in FIG. 13 will be described according to the flowchart shown in FIG.

  As a recommended content determination method, the content having the highest degree of preference for each part is determined as the content recommended to the viewer A every 10 minutes, and the reason for determining the recommended content is the preference of the content. This is the preference information of the viewer A that has the most influence on the degree. Further, the threshold value to be compared with the preference level of each content in the preference level comparison unit 110 is 0.

  Here, the recommended content determination process at 7:00 will be described.

  In step S401, since “content A”, “content B”, and “content C” are currently being received, the process proceeds to step S402.

  In step S402 to step S404, since the threshold value is 0, as the recommended content candidate, the partial name “A-1” of “content A”, the partial name “B-1” of “content B”, and “content C” ", The partial name" C-1 ".

  In step S405, the recommended content determination unit 111 determines whether there is a candidate for recommended content. Here, since there are three contents as described above, the process proceeds to step S406.

  In step S406, in order to determine the content with the highest degree of preference as the content recommended to the viewer A, the recommended content determination unit 111 determines the recommended content to be the partial name “A-1” of “content A-1”. The preference level of the partial name “A-1” of “content A-1” is increased by 13 due to the match between the performer “Mr. A” and the preference information, and the keyword “K1” and the preference information As a result of the match, the preference level is increased by 15 to reach 28 in total. Therefore, it is determined that the reason for determining the partial name “A-1” of “content A-1” as the recommended content is due to the keyword “K1”.

  Hereinafter, similarly, the recommended content determination process is performed every 10 minutes from 7:00 to 8:00, and the result of the process is shown in FIG.

  In FIG. 14, according to the time flow 640, the content 641 recommended to the viewer A every 10 minutes and the reason 642 for determining the recommended content 641 are displayed in association with the recommended content 641. For example, 10 minutes from 7:00 to 7:10 indicates that “content A” is recommended to the viewer A and is the largest reason that the keyword “K1” has been determined.

  However, the recommended content determination method for determining the recommended content is not necessarily limited to this.

  Next, the operation when the content display processing unit 115 outputs the recommended content shown in FIG. 14 determined based on the degree of preference for each part of each content shown in FIG. 13 to the display device 16 is shown in FIG. A description will be given according to the flowchart of the display process.

  Here, a case where the recommended content is changed from “content A” to “content B” at 7:30 will be described. As the output format, the reason for determining the recommended content and the preference level of the recommended content are displayed in the upper right corner, and the displayed content is forcibly switched to the recommended content.

  In step S451, the output processing unit 114 determines the output format of the recommended content. Here, in order to forcibly switch the displayed content and output “content B” as the recommended content, the process proceeds to step S452.

  In step S452, the output processing unit 114 extracts “content B” from the content reproduced in the content reproduction processing unit 113, and the process proceeds to step S454.

  In step S454, the output processing unit 114 determines whether to output the reason for determining the recommended content. Here, in order to output the reason why the content to be recommended has been determined to be “content B”, the process proceeds to step S455.

  In step S455, the output processing unit 114 extracts the reason why the “content B” illustrated in FIG. 14 is recommended from the recommended content determination unit 111, and the process proceeds to step S456.

  In step S456, the output processing unit 114 determines whether to output the content preference level. In order to output the preference level of “content B”, the process proceeds to step S457.

  In step S457, the output processing unit 114 extracts the preference level of the partial name “B-4” illustrated in FIG. 13 from the preference level calculation unit 109, and the process proceeds to step S458.

  In step S458, the output processing unit 114 determines whether or not to output a preference level transition of the “content B” that is the recommended content. Since no data is output here, the process proceeds to step S460.

  In step S460, the output processing unit 114 determines whether there is a recommended content other than “content B”. Here, since only “content B” is recommended, the process proceeds to step S461.

  In step S461, the output processing unit 114 reads the output format described above, and proceeds to step S462.

  In step S462, in accordance with the read output format, the output processing unit 114 outputs “content B”, the reason why “content B” is recommended, and the preference level of “content B”, and the process proceeds to step S463.

  In step S463, the content display processing unit 115 determines whether or not to display the output in step S462 on the display device. Here, since it displays on a display apparatus, it progresses to step S464.

  In steps S464 and S465, the content display processing unit 115 displays the output in step S462 on the display device, and ends the process.

  FIG. 15 illustrates a state in which “content B”, the reason for recommending “content B”, and the preference level of “content B” are output to the display device 16.

  In FIG. 15, the content is output on the display device 660. Until 7:30, “content A” is output as shown in the upper diagram of FIG. 15, but when it reaches 7:30, it is forcibly switched to “content B” as shown in the lower diagram. At this time, the reason 661 for recommending “content B” to the viewer A and the preference level 662 of “content B” are displayed in the upper right corner of the screen.

  As described above, in the content recommendation device according to the present embodiment, according to the designated output format, the “content B” as the recommended content, the reason for recommending “content B”, and the preference level of “content B” are set. It can be output to the viewer A.

  Note that the reason 661 for recommending “content B” to the viewer A and the preference 662 for “content B” are not necessarily output, and only one of them may be output, or both. It does not matter if it is not output. Also, other different information may be output.

  Further, as an output format different from the output format forcibly switching to “content B” recommended at 7:30 shown in FIG. 15, for example, the displayed content remains as it is and only the recommended content name is displayed. May be presented to the viewer A.

  FIG. 16 shows how the screen changes in the display device 16 in this case.

  In FIG. 16, content is output on the display device 680. Until 7:30, only “content A” is output as shown in the upper diagram of FIG. On the other hand, at 7:30, as shown in the figure below, although the screen remains “content A”, a content 681 “content B is recommended” is displayed in the upper right corner of the screen. Thereby, it is possible to inform the viewer A that “content B” is currently recommended.

  Thus, the viewer A can be notified of the presence of the recommended content according to the designated output format.

  In FIG. 16, only the recommended content name is displayed, but information indicating the reason for recommending the recommended content to the viewer may be further displayed. Moreover, you may display further the information which shows the preference degree of recommended content.

  As another output format, the preference level of all contents may be presented to the viewer A.

  The state of the screen on the display device 16 in this case is shown in FIG.

  In FIG. 17, “content A” is displayed on the display device 700 at 7:15. At this time, the preference level 701 of all contents can be output to the upper right corner of the screen, so that the viewer A can be notified of the preference level of all contents. However, what is displayed is not limited to the preference level.

  Further, as another output format, information indicating the temporal transition of the degree of preference for the content may be presented to the viewer A. Note that the information indicating the temporal transition of the degree of preference for the content may be displayed numerically or may be displayed in a graph form.

  The state of the screen on the display device 16 in this case is shown in FIG.

  In FIG. 18, “content A” is displayed on the display device 720 at 7:05. At this time, the graph 721 indicating the temporal transition of the degree of preference of “content A” from 7 o'clock to 8 o'clock is output in the upper right corner of the screen, thereby allowing the viewer A to temporally calculate the degree of preference of “content A”. Can inform the transition. However, not only “content A” but also the temporal transition of the preference level of all contents may be presented to the viewer A.

  In FIG. 18, only the temporal transition of the content preference level is displayed, but information that can identify the recommended content or the recommended content such as the recommended content name may be further displayed.

  As another output format, when the display device 16 has a screen division function, a plurality of contents may be output by adjusting the area of the screen to be divided according to the degree of preference.

  FIG. 19A shows five contents that are currently broadcast, and FIG. 19B shows that when these five contents exist, the display device 722 responds to the degree of preference of each content. The screen is divided into a plurality of areas, and the contents are output so that the respective contents are displayed in the divided areas corresponding to the degree of preference.

  In FIG. 19A, the content name 720 currently broadcast and the viewer's preference 721 for the content at a certain time are displayed in association with the content. As shown in FIG. 19B, the content display processing unit 115 divides the screen of the display device 722 into five areas according to the degree of preference of these five contents, and the largest divided area. “Content D” having a preference level of 50 is displayed in 723, “Content E” having a preference level of 30 is displayed in the second largest divided area 724, and “Content F” having a preference degree of 20 is displayed in the third largest divided area 725. And “contents G” and “contents H” having a preference level of 5 are displayed in the smallest divided areas 726 and 727, respectively.

  However, the present invention is not limited to displaying all contents. For example, when contents are arranged in order of preference, only the top two contents may be displayed using the screen division function, or only the top three contents may be displayed using the screen division function.

  Furthermore, the present invention is not limited to dividing the screen into a plurality of areas according to the degree of preference. For example, the content may be displayed by dividing the screen into a plurality of regions in equal divisions.

  Here, as described above, the five output methods of FIGS. 15 to 19 are shown, but the present invention is not necessarily limited to these.

  As described above, in the configuration of the present embodiment, the content preference level is calculated for each part in the currently broadcast content, and the content recommended to the viewer is determined based on the calculated preference level. The recommended content can be presented to the viewer. When calculating the preference level of each content, the preference information can be weighted to calculate the preference level. Moreover, as an output method to the display device, not only the recommended content is forcibly output to the screen, but only the recommended content name is output without changing the display screen, or the preference level of all content or the preference level of the content You can display the temporal transition of Furthermore, when the display device has a screen division function, the content is displayed by dividing the screen into a plurality of areas in equal divisions, or by dividing the screen into different sizes according to the degree of preference. Can also be displayed.

(Example 2)
In the second embodiment described below, an operation in the case where the preference level of the content is calculated at predetermined time intervals in the currently broadcast content will be exemplified.

  Here, it is assumed that the viewer A's preference information 600 shown in FIG. 12 and the viewing history information of the viewer A shown in FIG.

  FIG. 20 shows the contents of the viewer A in the past at which time zone regarding the “contents A”, “contents B”, and “contents C” broadcast from “7:00” to “8:00”. The viewing contents 741 of the viewer A for each time zone are shown according to the time flow 740. For example, “Content A” is viewed from “7:00” to “7:10”, and no content is viewed from “7:10” to “7:20”. To “7:30” indicates that “content A” is being viewed.

  The operation of the content recommendation device 100 in this embodiment will be described with reference to the flowchart shown in FIG. Here, in the state where “content A”, “content B”, and “content C” are broadcast with the contents as shown in FIG. 8, when the viewer A views the content from “7:00” The content recommendation device 100 calculates a preference level for each predetermined time of each content. Note that the predetermined time is 10 minutes, and the preference level of each content is calculated every 10 minutes.

  In step S541 shown in FIG. 9, the viewer information extraction unit 107 determines whether or not viewer information relating to the viewer A exists in the viewer information storage unit 105.

  Since the viewer A's preference information 600 shown in FIG. 12 and the past viewing history information of the viewer A shown in FIG. 20 are managed in the viewer information storage unit 105, the process proceeds to step S542. In step S542, the viewer information extraction unit 107 reads the viewer A's preference information 600 shown in FIG. 12 and the viewer A's past viewing history information shown in FIG.

  In step S543, it is determined whether or not the preference levels of the three contents “content A” to “content C” have been calculated.

  In step S544, the content information extraction unit 108 reads the content information shown in FIG.

  In step S545, the preference level calculation unit 109 calculates the preference level of each content.

  Here, in step S545, as a method by which the preference level calculation unit 109 calculates the preference level, the preference information 600 of the viewer A is compared with each piece of partial information held by each portion. Assume that the degree of preference is increased by ten. In addition, the preference level is calculated in consideration of the weight assigned to each item. Furthermore, when the viewer A views the content at the predetermined time using the past viewing history information of the viewer A shown in FIG. 20, the preference level of the content is increased by 10. FIG. 21 shows the calculation result of the preference degree for each predetermined time based on such rules.

  Here, since the preference level of each content is calculated every 10 minutes, in “Content A” from 7:00 to 7:10, “Mr. A” as a performer matches the preference information 600 of the viewer A. Therefore, the weighting of 1.3 is performed and the preference level is increased by 13. Furthermore, since “K1” in the keyword matches the preference information 600 of the viewer A, 1.5 is weighted to increase the preference level by 15. Furthermore, as shown in FIG. 20, since the portion from “7:00” to “7:10” of “content A” has been viewed by the viewer A in the past, the preference level is increased by 10. Therefore, the preference level of “content A” from 7:00 to 7:10 is 38.

  Similarly, from 7:30 to 7:40 of “content B”, since “genre Y” in the genre matches the preference information 600 of the viewer A, 0.5 is weighted and the preference is increased by 5 . Furthermore, since “Mr. F” as a performer matches the preference information 600 of the viewer A, weighting 0.8 is performed and the preference level is increased by 8. Furthermore, since “J” as a performer matches the preference information 600 of the viewer A, weighting 1.1 is performed and the preference level is increased by 11. Further, since “K6” as the keyword matches the preference information 600 of the viewer A, weighting 0.6 is performed and the preference level is increased by 6. Further, since “K7” as a keyword matches the preference information 600 of the viewer A, weighting 0.8 is performed and the preference level is increased by 8. Therefore, the preference level from “content B” 7:30 to 7:40 is 38.

  Similarly, the processing from step S543 to step S545 is repeated, and the preference levels of all contents are calculated.

  In FIG. 21, according to the time flow 760, the preference degree 764 for each of “Content A” 761, “Content B” 762, and “Content C” 763 is shown every 10 minutes. For example, the preference level of “content A” from 7:00 to 7:10 is 38, and the preference level of “content B” from 7:00 to 7:10 is 27.

  The predetermined time is not limited to 10 minutes, and may be, for example, 5 minutes or 15 minutes.

  Further, the method for calculating the degree of preference of each content every predetermined time is not limited to this.

  As described above, in the configuration of the present embodiment, the content preference level is calculated every predetermined time among the currently broadcast content, and the content recommended to the viewer is determined based on the calculated preference level. The recommended content can be presented to the viewer.

(Example 3)
In Example 3 shown below, when the preference level of all the currently broadcasted content received by the receiving unit 102 does not exceed a predetermined threshold, the content recommendation device recommends content that is not currently broadcasted. It illustrates about. In this embodiment as well, the degree of preference is calculated for each part or every predetermined time for content that is currently being broadcast and can be received by the receiving unit 102.

  At a point in time when there is content that is currently being broadcast and can be received by the receiving unit 102 (hereinafter referred to as real-time content) and content that is not currently being broadcast but is stored in the content storage unit 106 (hereinafter referred to as non-real-time content). 22 (a) and 22 (b) show the degree of preference.

  In FIG. 22A, there are five real-time contents 780, and a content name 781 and a preference level 782 at a certain point in time for the contents are displayed in association with the contents. For example, the preference level at a certain point in time for “content J” currently being broadcast is 24.

  Similarly, in FIG. 22B, there are five non-real-time contents 783, and a content name 784 and a preference 785 at a certain point in time for the contents are displayed in association with the contents. For example, the preference level at a certain point in time for “content P” is 58.

  In addition, although the calculation method of a preference degree may be calculated by the method shown in Example 1 and Example 2, for example, it is not necessarily limited to these.

  The operation of the content recommendation device 100 in this embodiment will be described with reference to the flowchart shown in FIG. Here, as a recommended content determination method, when the preference level of all real-time content is less than or equal to a predetermined value (in this embodiment, 30 or less), the content with the highest preference level among non-real-time content is recommended content. As a recommendation to the viewer.

  Note that it is assumed that the five real-time contents and the five non-real-time contents shown in FIGS. 22A and 22B are reproduced by the content reproduction processing unit 113 so that the reproduction time zones overlap. Each content preferably has the same playback start time, but the playback start time may not be the same as long as at least the playback time zones overlap.

  In step S401, the recommended content determination unit 111 determines whether the content is currently received content (real-time content). Here, if the content is five contents (real-time content) from “content J” to “content N” shown in FIG. 22A, the process proceeds to step S402, and the content is “content” shown in FIG. In the case of five contents (non-real-time contents) from “P” to “content T”, the process proceeds to step S404.

  In step S <b> 402, the preference level comparison unit 110 compares the preference level of the content with the threshold 30. If the preference level of the content is 30 or more, the process proceeds to step S403, and if not, the process proceeds to step S404. Here, as shown in FIG. 22A, since all the five contents from “content J” to “content N” have a preference degree of 30 or less, the process always proceeds to step S404.

  In step S404, the recommended content determination unit 111 determines whether all content has been checked.

  In step S405, the recommended content determination unit 111 determines whether there is a candidate for recommended content. Here, since the process has not proceeded to step S403 as described above, there is no recommended content candidate. Therefore, the process proceeds to step S407.

  In step S407, as described above, since the real-time content preferences are all 30 or less, the content with the highest preference among non-real-time content is recommended to the viewer as recommended content. Therefore, the recommended content determination unit 111 recommends “content Q” having the highest preference among non-real-time content to the viewer A as the recommended content.

  Note that the threshold value used in step S402 is not necessarily limited to one that is uniquely determined. You may change according to a viewer, or may change according to the time zone which appreciates a content.

  The recommended content determination unit 111 determines the non-real-time content as the recommended content when all the real-time content preferences are equal to or less than the threshold, but determines the content with the highest preference among the non-real-time content as the recommended content. It is not limited to the structure to do. For example, by recording the number of times viewers have viewed non-real-time content in the past, when the preference level of real-time content is less than or equal to the threshold value, the most preferred content among the contents with the number of viewing times equal to or less than a predetermined threshold value The content with a high degree may be determined as the recommended content.

  In the configuration of this embodiment, the content recommendation device can recommend non-real-time content when the preference level of all the content currently being broadcast received by the receiving unit 102 does not exceed a predetermined threshold. Thus, not only real-time content but also non-real-time content, the degree of preference is calculated for each part or every predetermined time, and by using the calculated degree of preference, the most viewer at that time Can be presented to the viewer.

  In the above description, the preference level calculation unit 109 calculates the preference level for each predetermined time, calculates the preference level within a predetermined time every predetermined time, or sets the preference level at that time for every predetermined time. Although described as an operation to calculate, in each example, any of these preference level calculation operations can be adopted.

  As described above, according to the content recommendation device of the present invention, the viewer information including the viewer attributes and the viewer preference information and the content information attached to the content are used at predetermined time intervals or at the content. By calculating the preference level for the same content multiple times for each constituent part and determining the content to recommend to the viewer each time, it is more accurate than calculating the preference level for each content. Can provide content that suits viewers' preferences. Accordingly, it is possible to appropriately determine the degree of preference of the viewer at the time of viewing the program and to recommend the content reflecting the degree of preference that changes as the program content changes.

  In addition to outputting recommended content that suits the viewer's preference, it also outputs the degree of preference for the content and the reason for recommending the content, so that the viewer has output the content. Can know.

  In addition, when a display device is provided, not only the recommended content for the current viewer but also the recommended content name is displayed to the viewer, or the preference information of other content is displayed. be able to. Furthermore, when the display device has a screen division function, the screen can be divided into a plurality of regions according to the degree of preference of the content, and each content can be distributed and displayed in the divided regions.

  The present invention is not limited to the above-described embodiments, and various modifications are possible within the scope shown in the claims, and embodiments obtained by appropriately combining technical means disclosed in different embodiments. Is also included in the technical scope of the present invention.

  Since the present invention can recommend content to the viewer based on the preference calculated every predetermined time, the present invention is also applicable to devices having a content recording function or a content playback function, such as a hard disk recorder. it can.

It is a block diagram which shows the functional structure of the content recommendation apparatus which concerns on one embodiment of this invention. It is a block diagram which shows the structure of the computer system which implement | achieves each function part of the content recommendation apparatus shown in FIG. It is a flowchart which shows the main operation | movement of the content recommendation apparatus shown in FIG. It is a flowchart which shows the preference degree calculation process performed by step S304 of FIG. It is a flowchart which shows the recommendation content determination process performed by step S305 of FIG. It is a flowchart which shows the output / display process performed by step S307 of FIG. It is explanatory drawing which shows the viewer A's preference information regarding one Example of the content recommendation apparatus shown in FIG. With respect to one embodiment and other embodiments of the content recommendation apparatus shown in FIG. 1, detailed contents of each part constituting three contents broadcasted from “7:00” to “8:00” are shown. It is explanatory drawing shown. 5 is a flowchart illustrating an operation of the content recommendation device when calculating the degree of preference of each content with respect to the embodiment of the content recommendation device shown in FIG. 1. It is explanatory drawing which shows the result at the time of calculating a preference degree for every part which comprises each of three content regarding one Example of the content recommendation apparatus shown in FIG. It is explanatory drawing which shows the result at the time of calculating a preference degree for every part which comprises each of three content regarding one Example of the content recommendation apparatus shown in FIG. It is explanatory drawing which shows the viewer's A preference information and the weighting value of each item regarding one Example of the content recommendation apparatus shown in FIG. It is explanatory drawing which shows the result at the time of calculating a preference degree for every part which comprises each of three content regarding one Example of the content recommendation apparatus shown in FIG. It is explanatory drawing which shows the reason which came to determine the recommended content and the said recommended content every 10 minutes regarding one Example of the content recommendation apparatus shown in FIG. It is explanatory drawing which shows an example of the display form of the screen according to an output format regarding one Example of the content recommendation apparatus shown in FIG. It is explanatory drawing which shows the other example of the display form of the screen according to an output format regarding one Example of the content recommendation apparatus shown in FIG. It is explanatory drawing which shows the further another example of the display format of the screen according to an output format regarding one Example of the content recommendation apparatus shown in FIG. It is explanatory drawing which shows the further another example of the display format of the screen according to an output format regarding one Example of the content recommendation apparatus shown in FIG. FIG. 19 (a) is an explanatory diagram showing the preference level at a certain point in time for the five contents currently broadcast, and FIG. 19 (b) relates to one embodiment of the content recommendation apparatus shown in FIG. It is explanatory drawing which shows an example of the display form of the screen according to an output format in case one content exists. It is explanatory drawing which shows the viewing history of the past content of the viewer A regarding the other Example of the content recommendation apparatus shown in FIG. It is explanatory drawing which shows the result at the time of calculating a preference degree every 10 minutes in each of three content regarding the other Example of the content recommendation apparatus shown in FIG. FIG. 22 (a) relates to still another embodiment of the content recommendation device shown in FIG. 1, and FIG. 22 (b) is a diagram illustrating the preference of real-time content currently being broadcast. It is explanatory drawing which shows the preference degree of real-time content. FIG. 23 (a) is an explanatory diagram for explaining processing for calculating the preference degree at every predetermined time in the content recommendation device shown in FIG. 1, and FIG. 23 (b) shows the preference degree for each part in the content recommendation device. It is explanatory drawing explaining the process to perform. FIG. 3 is an explanatory diagram showing a state when a preference level is calculated from content information and viewer information in the content recommendation device shown in FIG. 1.

Explanation of symbols

DESCRIPTION OF SYMBOLS 11 Viewer information 12 Content / content information 14 Viewer identification information 16 Display apparatus 100 Content recommendation apparatus 102 Receiving part (reception means)
105 viewer information storage unit (viewer information storage means)
106 Content storage unit (content storage means)
109 Preference level calculation unit (preference level calculation means)
110 Preference level comparison unit (comparison means)
111 Recommended content determination unit (recommended content determination means)
113 Content reproduction processing unit (content reproduction means)
114 Output processing unit (output means)
115 Content display processing unit (display control means)

Claims (22)

Viewer information storage means for storing viewer information including information on individual viewers;
Content storage means for storing content and content information as information related to the content;
Content reproduction means for reproducing one or more contents stored in the content storage means;
A preference level calculation unit that calculates the preference level a plurality of times using the viewer information and the content information of the content for the same content being played back by the content playback unit;
Each time the preference level for the content currently being played back is calculated by the preference level calculation unit, the content is provided with a recommended content determination unit that refers to the calculated preference level and determines the content recommended for the viewer. A content recommendation device.   2. The content recommendation device according to claim 1, further comprising output means for outputting recommended content for the viewer determined by the recommended content determination means.   2. The content recommendation device according to claim 1, further comprising output means for outputting information capable of specifying recommended content for the viewer determined by the recommended content determination means.   4. The content recommendation device according to claim 2, wherein the output unit further outputs information indicating a reason for recommending a recommended content to the viewer determined by the recommended content determination unit. .   4. The content recommendation apparatus according to claim 2, wherein the output unit further outputs information indicating a degree of preference of the recommended content to the viewer determined by the recommended content determination unit.   In the case where a plurality of contents are reproduced by the content reproduction means, an output means is provided for outputting information indicating the preference degree calculated by the preference degree calculation means for these contents. The content recommendation device according to claim 1.   3. The content recommendation device according to claim 2, wherein the output unit outputs information indicating a transition of the preference level calculated by the preference level calculation unit for the recommended content.   The output means, when a plurality of contents are played back by the content playback means, outputs information indicating a transition of the preference level calculated by the preference level calculation means for these contents. The content recommendation device according to claim 2.   The preference level calculation unit calculates a preference level for the same content every predetermined time, and the recommended content determination unit determines a content recommended for a viewer every predetermined time. The content recommendation device according to claim 1. The content is composed of a plurality of parts,
The preference level calculation unit calculates a preference level for each part of the same content, and the recommended content determination unit determines a content recommended for a viewer for each part. The content recommendation device according to 1.   The content stored by the content storage means is at least one of content currently being broadcast, content broadcast in the past, content read from an external recording medium, or content downloaded via a network. The content recommendation device according to claim 1, wherein the content recommendation device is provided.   The viewer preference information included in the viewer information includes the viewer's favorite genre, the viewer's favorite person, the viewer's favorite keyword, the viewer's favorite content, and the viewing. The content recommendation device according to claim 1, wherein the content recommendation device is at least one of viewing history information of past contents of the user. Receiving means for receiving content;
Comparing means for comparing the preference level calculated by the preference level calculation means with a predetermined threshold,
The recommended content determination means, when the comparison means determines that the preference level of the content currently receivable by the reception means does not exceed the predetermined threshold, the content reproduction means The content recommendation device according to claim 1, wherein content having the highest preference is recommended from second content other than the first content currently receivable by the receiving unit.   The recommended content determination unit, when determining recommended content from second content, determines recommended content from content whose past number of times of viewing is equal to or less than a predetermined threshold. Content recommendation device. Comparing means for comparing the preference level calculated by the preference level calculation means with a predetermined threshold,
2. The content recommendation apparatus according to claim 1, wherein the recommended content determination unit excludes content whose preference level does not exceed the threshold from recommended content candidates.   16. The content recommendation device according to claim 15, wherein the comparison unit changes the threshold according to a viewer.   16. The content recommendation device according to claim 15, wherein the comparison unit changes the threshold according to a time period during which a viewer appreciates the content. Comparison means for comparing the preference level calculated by the preference level calculation means with a predetermined threshold;
Display control means for displaying on the display device the information output from the output means,
The display control unit displays the screen of the display unit when the comparison unit determines that the preference level of the plurality of contents exceeds a predetermined threshold among the plurality of contents reproduced by the content reproduction unit. The content recommendation apparatus according to claim 2, wherein a plurality of contents having a preference degree exceeding a predetermined threshold are distributed and displayed in the divided areas.   9. The content recommendation device according to claim 2, further comprising display means for displaying information output by the output means. Viewer information storage step for storing viewer information including information about individual viewers;
A content accumulation step for accumulating content and content information as information relating to the content;
A content playback step of playing back one or more contents stored in the content storage step;
A preference level calculating step of calculating a preference level a plurality of times using the viewer information and the content information of the content for the same content being played back in the content playback step;
And a recommended content determination step of calculating the calculated preference level and determining the content recommended to the viewer each time the preference level for the content currently being played back is calculated by the preference level calculation step. Content recommendation method.   A program for operating the content recommendation device according to any one of claims 1 to 19, wherein the content recommendation program causes a computer to function as each of the above means.   A computer-readable recording medium on which the content recommendation program according to claim 21 is recorded.

Images