JP5237849B2 - Program recommendation device - Google Patents

Description

  The present invention relates to an apparatus for recommending a program that matches a user preference to a user.

  In the modern information civilized society, the role of broadcasting is immeasurable. In particular, the influence of television broadcasting on viewers is great. The penetration rate of television is extremely high, and it is installed in almost all households. Broadcast content distributed from each broadcast station is viewed by an unspecified number of people.

  Recently, with the development of digital technology, it has become possible to store a large amount of video content, and HDD-based recorders and personal computers with TV program recording / viewing functions have appeared.

  Furthermore, with the advent of server-type distribution systems such as VOD (Video On Demand), users can view broadcast programs as much as they like at any time. In addition, as the number of broadcast channels has increased, users have been required to select their favorite programs from a large number of programs.

  For this reason, a technique for recommending a program that matches each user's preference is actively researched as an apparatus that supports the user's program selection.

  The most important thing in the program recommendation technology is to appropriately prepare user preference information (what the user is interested in) and use the user's behavior / operation history to obtain the user's preference information. There is research to produce.

  In Patent Document 1, user preference information is extracted using a program control operation history of a user's remote controller, and program recommendation is performed based on the extracted user preference information. For example, the program attention level of the user is calculated by an operation of volume adjustment using a remote controller.

  In Patent Literature 2, preference information is generated by adding a score to a keyword included in a program reserved by the user by using the program reservation history of the user.

In Patent Document 3, the program genre that the user is interested in is recommended by analyzing the appearance frequency of the program genre viewed by the user.
In Patent Document 4, a mobile phone is used as a remote controller for an electric device.
Patent Document 5 proposes a mobile phone with a remote control function capable of selecting and switching a reception channel.

JP 2002-369090 A JP 2006-311010 A JP 2008-131339 A JP 2002-204292 A JP 2005-150831 A

  However, in the present age when computerization accelerates, users can always send and receive information using various terminals, and users are increasingly using multiple terminals at the same time. As a result, the user often uses a plurality of terminals even while viewing the program, and does not always concentrate on viewing the program.

  Therefore, in the conventional method, the keyword included in the program viewed by the user is regarded as information that the user is interested in, and it is determined how concentrated the user has viewed the program (concentration level). There is a problem that it is not possible.

  The present invention has been made in view of the above-described points, and the object of the present invention is to accurately extract the degree of concentration of each user, appropriately generate each user's preference information, and respond to each user's preference. The recommendation is to recommend matching programs.

  In order to achieve the above object, the program recommendation device according to the present invention determines how concentrated the user has viewed the program by acquiring a history of operations other than program control and program control, and The degree of concentration on the user's program is extracted.

The invention described in claim 1 is a program recommendation device that performs program recommendation based on user preference ,
An operation history acquisition unit that acquires the user's terminal operation history from the user's terminal for operations other than program control operation and program control ;
A viewing history extraction unit that extracts the user's program viewing history based on the terminal operation history of the program control operation acquired by the operation history acquisition unit ;
And program viewing history of the user that the viewing history extraction part is extracted, and generating preference information of the user based on said user terminal operation history of the operation other than the program control the operation history acquiring unit acquires, A program evaluation unit that records and stores the user preference information in program evaluation metadata ;
Topic / keyword acquisition means for acquiring global topics and / or trendy keywords,
Other user preference information acquisition means for acquiring the other user's preference information stored in the other user's program recommendation device;
Recommending a program selected based on popular topics and / or popular keywords acquired by the topic / keyword acquisition means when the number of times the user views the program is less than a first predetermined number of times; The user preference information recorded and stored in the program evaluation metadata when the number of times of program viewing is equal to or more than the first predetermined number of times and less than a second predetermined number of times greater than the first predetermined number of times When the program selected based on the user is recommended and the number of times the user views the program is equal to or greater than the second predetermined number, the other user's preference information acquired by the other user's preference information acquisition means and the program evaluation meta By comparing the user's preference information recorded and stored with data and extracting the viewing history of the other user that is similar to the user, the similar user Said Although viewing user is one that still have a and the program recommendation unit configured to recommend a program that is not watching.

The invention described in claim 2 is the program recommendation device described in claim 1 ,
It further has program search means for searching for a program based on the user input information.

The invention described in claim 3 is the program recommendation device described in claim 2 ,
The program search unit is configured to perform a program search using the preference information of the user stored in the program evaluation metadata.

The invention described in claim 4 is the program recommendation device described in claim 1 ,
A program storage unit for storing the recorded program is further provided.

The invention described in claim 5 is the program recommendation device according to claim 1 ,
Terminal the operation history acquiring unit acquires the operation history has a function of switching a program control function and a terminal original function before SL terminal operation history acquiring unit, obtains the operation history of the terminal original function It is characterized by.

The invention described in claim 6 is the program recommendation device according to claim 1 ,
The program evaluation unit is configured to generate the preference information of the user according to the operation content by the terminal of the user.

  According to the feature of the present invention, the user can select a program that matches his / her preference.

  Moreover, according to the structure which identifies a some user, the effect that the program which matched preference for every user and every group can be selected is acquired.

It is a figure explaining the structure of one Example of this invention. It is a figure explaining the flow of processing of a program recommendation device. It is a figure explaining the example of program schedule data. It is a figure explaining the example of a user's program viewing history and a terminal operation history. It is a figure explaining the example of how to add the evaluation score of a user's preference information. It is a figure explaining the example of user preference information. It is a figure explaining the example of the user preference information accumulate | stored in program evaluation metadata. It is a figure explaining the method of the recommendation processing method switching by the learning step of preference information. It is a figure explaining the method a user selects the recommendation processing method. It is a figure explaining the method of controlling preference information with the operation content of a terminal. It is a figure explaining the example of program schedule data. It is a figure explaining the example of the calculation method of preference information. It is a figure explaining the example of the calculation method of preference information using a dynamic operation history. It is a figure showing the structure where a some user exists in a program recommendation apparatus. It is the flow of processing when there are multiple users. It is a performer evaluation score table and a genre evaluation score table in the case of a plurality of users. This is an example of grouping when a plurality of users view simultaneously.

  Embodiments of the present invention will be described below with reference to the drawings.

  FIG. 1 is a diagram illustrating the configuration of the first embodiment. In the present invention, the terminal that performs the program control operation and the terminal that performs an operation other than the program control may not be the same terminal. In the first embodiment, it is assumed that the terminal has a function capable of performing both operations. Here, the term “terminal” refers to a device that has a user interface and communication means and has a function of controlling other devices.

  The terminal 110 includes an operation instruction unit 111 and a communication unit 112. The program recommendation device 120 includes an operation history acquisition unit 121, a viewing history extraction unit 122, a program evaluation unit 123, a program evaluation metadata 124, and a program recommendation unit 125, which are, for example, set-top boxes (connected to a television, cable TV broadcast, satellite broadcast, broadband VOD broadcast, communication karaoke, Web browsing, and other video content receivers such as television). As the program evaluation metadata 124, for example, an embedded database (a lightweight, high-speed, high-function relational database that can be used in an embedded system) can be used. The video distribution server 130 is connected to the program recommendation device 120 via a network. The program evaluation metadata 124 may be in the video distribution server 130 or shared with other users. It is also possible to use an embedded database in the program recommendation device 120 and an enterprise database on the video distribution server 130 side by side. The display unit 140 is a television or a monitor, and may be built in the terminal 110 or the program recommendation device 120.

Hereinafter, each part which comprises a present Example is demonstrated.
The terminal 110 is a mobile terminal having a program control function, and a mobile phone or the like can be applied.
The operation instruction unit 111 is an input unit used for a user to perform an operation by button input. The operations performed by the user include operations other than switching between program control operations and other operations, program control operations, and program control operations. The program control operation is an operation including a channel change. When the terminal 110 is a mobile phone, a mobile application necessary for the operation is started and operated. An operation other than program control means an operation other than program control, such as mail transmission / reception, telephone call, and web operation, when the terminal 110 is a mobile phone.
The communication unit 112 is a unit that enables communication with the program recommendation device 120 by infrared communication or the like. Communication using BlueTooth (registered trademark) can also be used.

  The operation history acquisition unit 121 starts acquisition with the user's viewing start / end information as an input. When the user changes the program using the terminal 110, the information is acquired by the operation history acquisition unit 121. The operation history acquisition unit 121 acquires an operation history of the mobile terminal 110 that is a user's viewing terminal and is turned on. The viewing history extraction unit 122 extracts program information and viewing time viewed by the user from the user's program control operation history acquired by the operation history acquisition unit 121. As the viewing program information, electronic data of program information such as program title, genre, performer name, and program outline description is used. Such data is obtained from the video distribution server 130 via a network, for example. Specifically, it is electronic data of program information such as a program title, a genre, a performer name, and a program outline description. The operation history acquired by the operation history acquisition unit 121 targets operations other than program control and program control of the terminal 110. The operations other than the program control are, for example, mail transmission / reception, telephone calls, and web operations in a mobile phone.

The program evaluation unit 123 receives the program viewing history and the portable terminal operation history as input, calculates the user's degree of concentration on the program, and records / stores it in the program evaluation metadata 124.
The program evaluation metadata 124 manages and manages the user preference information generated by the program evaluation unit 123 by holding or updating it.
The program recommendation unit 125 recommends an appropriate program to the user based on the user preference information accumulated in the program evaluation metadata 124.

FIG. 2 is a diagram illustrating a processing flow of the present embodiment.
In step 201, a program control operation using a mobile phone is started, and program viewing is started (step 202). This determination may be made by, for example, pressing a button for switching a program control operation and other operations on the mobile phone. The power supply of the viewing terminal may be used as the determination means. In step 203, acquisition of the operation history of the user's mobile phone is started. In step 204, when the operation button of the mobile phone is pressed, it is determined whether or not the pressed button is program control. When a button other than program control is pressed, the operation history is transmitted to the operation history acquisition unit 121, and the operation history acquisition unit 121 accumulates the received operation history. When the program control button is pressed, the type of the button (power OFF button, recommended button, or other) is further determined. When the recommendation button is pressed, the program recommendation unit 125 performs a program recommendation process and presents the recommended program on the display unit 140 (step 206). When a button that is neither recommended nor turned off is pressed, program control is executed, the operation history is transmitted to the operation history acquisition unit 121, and the operation history acquisition unit 121 accumulates the received operation history. If the power OFF button is pressed, the viewing history is extracted in step 208 using the operation history of the mobile phone program control accumulated in step 205. Then, the user's program preference information is generated by inputting two data of operation history and viewing history other than program control (step 209), and the preference information is registered and accumulated in the program evaluation metadata 124 (step 210). In step 211, the program guide data is updated from the video distribution server 130, and the process waits until the next program control operation is performed. The operation history may be accumulated by the mobile phone until the power is turned off. In this case, the operation history may be transmitted to the operation history acquisition unit 121 when the power is turned off. The program guide data may take a format that the user inputs to the program evaluation metadata 124 using a recording medium such as a USB.

Next, the processing method of program preference information extraction at step 209 and program recommendation at step 206 will be described.
As shown in FIG. 3, the program guide data is composed of a program name, performers, genres, program length, and the like. This is distributed from the video distribution server 130, for example.
The program evaluation score Qp of the viewed program P includes the program length Lp registered in the program guide data, the viewing time Xp of the program P input in steps 208 and 209, and the operation time Ypi of the mobile phone while viewing the program. The sum is calculated as shown in Equation 1 (FIG. 4).

Here, the program evaluation score Qp is defined so as to continuously take a value of 0 or more and 1 or less. However, any value may be used as long as the degree of concentration, that is, the program evaluation score is different. A discrete value may be taken.
When the program evaluation score Qp is calculated, the evaluation score Qp is added to those related to the program in the performer evaluation score table and the genre evaluation score table (FIG. 5). Here, the evaluation score table may be given to each user of the program recommendation device 120.

  The user program preference information generated as described above is shown in FIG. 6 and is stored in the program evaluation metadata 124. Here, evaluation scores are assigned only to performers and genres, but evaluation scores may also be assigned to keywords and the like in the program. The evaluation score of the program preference information may be deducted after a certain period of time. For example, for items that you did not watch, 0.1 points will be deducted daily. At this time, the evaluation score may not take a negative value, or may be deducted by 0.1 points every power-on day.

  Based on the user's program preference information stored in the program evaluation metadata 124, a score is given to the program guide data of FIG. 3 in step 206 (FIG. 7 (a)). Then, the recommendation is made in descending order, and the recommended program name is displayed on the display unit, and the user selects which program to view from the recommended programs. Here, as shown in FIG. 7 (b), the reason for recommendation, that is, the rating of the program is shown on the display, and the user inputs his / her own rating for the rating, and the user preference information is displayed. It may be possible to rewrite. Further, the user preference information may be stored as electronic data in the program evaluation metadata 124 and may be transferred to another medium using a recording medium such as a USB memory. It may be possible to input other users' preference information to the program recommendation device and perform recommendation processing using the preference information. Further, backup data may be left in the video distribution server 130.

  Next, the program recommendation process in step 206 will be described with reference to FIGS. In step 206, when the program recommendation process is started, the recommendation processing method is switched.

  In FIG. 8A, the recommendation processing method is switched depending on the number of times the user has viewed the program. The number of times the program is viewed may be calculated as the number of times the power is turned on, or may be the number of times the program has been viewed. If the number of times the program has been viewed is less than X times, the user's preference information has not been sufficiently accumulated. Therefore, popular topics and popular keywords are acquired from the Web, and related programs are recommended. A user may input a topic or popular keyword in the world. If the number of viewing times is X times or more and less than Y times, a program having a high score is recommended from the user's preference information accumulated as described above. If the number of viewing times is Y or more, the recommended program based on the user's preference information becomes uniform, so the preference information of the other user stored in the program recommendation device of the other user is compared with the preference information of the user. Then, by extracting viewing histories of similar users, it recommends programs that other similar users are viewing but have not yet viewed. The viewing times X and Y of such recommendation process switching may be input by the user or may be registered at the time of product purchase. An input example is shown in FIG. Further, the recommendation process may be switched by using a means for the user to select using the display unit as shown in FIG. It may be based on input of other external information such as input from the video distribution server side.

Next, step 205 will be described in detail with reference to FIG.
When a mobile phone operation other than program control is performed, the operation is classified according to the type of operation. The operations here are calls, mail transmission / reception, and web operations. These operations can be discriminated by a button on the mobile phone.
When a call / mail transmission / reception is performed, it is next determined whether it is transmitted or received. In the case of transmission, it is further determined whether or not a program-related word is included in the mail text. If a program related word is included, it is considered that the program is interested, and one point of the highest score is added as the evaluation score. If the program related word is not included, it is considered that the program is not interested, and the above-mentioned program evaluation score is used. In the case of reception, since the degree of concentration on the program is lowered, the program evaluation score is added in the same manner. Whether or not a program-related word is included in the mail is determined by the pressed button. You may perform using processes, such as character recognition.

  When web operation is performed, it is divided according to whether related keyword search is performed. If the related keyword search is not performed, the user is regarded as not interested in the program, and the program evaluation score of FIG. 4 is added. If a related keyword search is performed, it is considered that the program is interested, and one point of the highest score is added. Whether or not the related keyword search has been performed is determined based on the referred URL. Up to this point, the highest evaluation score is 1, but the highest evaluation score can be changed by the definition of Equation 1.

The above can be implemented when the program guide data uses an electronic program guide as shown in FIG.
Hereinafter, a program evaluation method when more detailed program guide data as shown in FIG. 11 is used will be described.

  FIG. 11 shows program guide data to which a timetable of a certain program is also assigned. Performers, genres, program heads, etc. in each time zone are given. In accordance with this, program viewing time and mobile phone operation time are also extracted for each time zone.

  FIG. 12 shows a system in which program evaluation scores are calculated for performers, genres, etc. in each time zone by using the extracted program viewing time and mobile phone operation time as input, and points are added. When the viewing time of the program P and the operation time of the mobile phone in the meantime are given as shown in FIG. 12, the program evaluation score in each time zone is given by the following equations (2) to (5).

  Equations 2 to 5 are obtained by applying Equation 1 to each time zone. By performing such a calculation, it is possible to acquire program preference information with high accuracy.

  In FIG. 13, performers and genres appearing in the time zone when the operation of the mobile phone has started are regarded as information that is not particularly interesting, and a negative evaluation score is added as shown in Equation 6.

  Further, performers and genres appearing in a time zone when the operation of the mobile phone is stopped are regarded as information of interest, and further points are added using the evaluation score of Equation 7.

  In the present embodiment, a portable terminal such as a mobile phone having a program control operation and an operation other than the program control is assumed as the terminal 110, but a terminal having a program control operation and a terminal having an operation other than the program control may be different. . Since it is assumed that operations other than program control are operations indicating that the program is not concentrated on viewing the program, such an operation may be an operation such as pressing an air conditioner switch, for example. Therefore, a home appliance such as an air conditioner or a microwave oven may be used as a terminal for performing operations other than program control.

Hereinafter, the case where Example 1 is expanded to Example 2 will be described.
FIG. 14 is a diagram illustrating a configuration in which a plurality of users exist in the program recommendation device in the first embodiment. The terminal 1410 and the terminal 1450 are the same as the terminal 110 in FIG. The program recommendation device 1420 has a user identification unit 1421 in addition to the configuration of FIG. In this configuration, there are two terminals, but three or more terminals may be used.

The user identification unit 1420 has a function of identifying a user using information such as a personal ID registered in each of the terminal 1410 and the terminal 1450. Communication means such as infrared rays are used.
Other configurations are the same as the configuration in FIG. 1 of the first embodiment.

  FIG. 15 shows the flow of processing when there are a plurality of users. Next, a difference from the process of FIG. 2 according to the first embodiment will be described.

  In step 1501, when a program control operation of the user's mobile terminal is started, in step 1502, the user is identified. The processing after identifying the user is the same as in the first embodiment, and the user preference information is generated for each user as shown in FIGS. At this time, the number of users who view the program is not limited to one. When a plurality of users view the program at the same time, grouping is performed to generate preference information as shown in FIG. The recommendation process is the same as in the first embodiment.

  In FIG. 14, a program search unit 1424 searches for a program from a keyword input by the user himself / herself from programs stored in the video distribution server 1430 or the program guide metadata 1424. At this time, keyword input is supported using user preference information stored in the program evaluation metadata 1425. For example, when one character at the beginning of a keyword is input, keywords having a high score starting with that character may be listed. Even if nothing is input, keywords having such a high score may be listed.

  The present invention is a video program for home video devices such as Blu-ray Disc players, DVD recorders, TVs, personal computers with TV functions, information devices such as PDAs and mobile phones, and set-top boxes that play online distribution programs. Applicable to devices to watch.

110 Terminal 111 Operation instruction unit 112 Communication means 120 Program recommendation device 121 Operation history acquisition unit 122 Viewing history extraction unit 123 Program evaluation unit 124 Program guide metadata 125 Program recommendation unit 130 Video distribution server 140 Display unit

Claims (6)

A program recommendation device for recommending a program based on user preferences,
An operation history acquisition unit that acquires the user's terminal operation history from the user's terminal for operations other than program control operation and program control ;
A viewing history extraction unit that extracts the user's program viewing history based on the terminal operation history of the program control operation acquired by the operation history acquisition unit ;
And program viewing history of the user that the viewing history extraction part is extracted, and generating preference information of the user based on said user terminal operation history of the operation other than the program control the operation history acquiring unit acquires, A program evaluation unit that records and stores the user preference information in program evaluation metadata ;
Topic / keyword acquisition means for acquiring global topics and / or trendy keywords,
Other user preference information acquisition means for acquiring the other user's preference information stored in the other user's program recommendation device;
Recommending a program selected based on popular topics and / or popular keywords acquired by the topic / keyword acquisition means when the number of times the user views the program is less than a first predetermined number of times; The user preference information recorded and stored in the program evaluation metadata when the number of times of program viewing is equal to or more than the first predetermined number of times and less than a second predetermined number of times greater than the first predetermined number of times When the program selected based on the user is recommended and the number of times the user views the program is equal to or greater than the second predetermined number, the other user's preference information acquired by the other user's preference information acquisition means and the program evaluation meta By comparing the user's preference information recorded and stored with data and extracting the viewing history of the other user that is similar to the user, the similar user Program recommendation apparatus, characterized by chromatic and program recommendation unit for recommending the user but is viewing is not yet watching the program. The program recommendation device according to claim 1 ,
The program recommendation device further comprising program search means for searching for a program based on the user input information. The program recommendation device according to claim 2 ,
The program search unit, the program recommendation apparatus characterized by performing the program search by using the preference information of the user stored in the program evaluation metadata. The program recommendation device according to claim 1 ,
A program recommendation device further comprising a program storage unit for storing recorded programs. The program recommendation device according to claim 1 ,
Terminal the operation history acquiring unit acquires the operation history has a function of switching a program control function and a terminal original function before SL terminal operation history acquiring unit, obtains the operation history of the terminal original function A program recommendation device characterized by the above. The program recommendation device according to claim 1 ,
The program evaluation unit, the program recommendation apparatus and generates the preference information of the user according to the operation content by the terminal of the user.

Images