JP2009015560A - List generation device and method, and computer program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To much more suitably present content suited to the conflicting tastes of a user. <P>SOLUTION: A list generation device is provided with first acquisition parts (11, 12, 13, 14) for acquiring a plurality of content parameters showing characteristics of each content; second acquisition means (21, 22, 23 and 24) for acquiring a plurality of user parameters showing characteristics of a user; an arithmetic means (31) for performing prescribed arithmetic processing by using a plurality of content parameters and a plurality of user parameters according to the association of a plurality of content characteristic items and a plurality of user characteristic items for each of a plurality of content; and a generation means (32) for generating a content list by listing a plurality of contents in the order that the content is suited to the tastes of the user based on the result of the arithmetic processing. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to, for example, a list generation apparatus and method for preferentially presenting to a user content that matches a user's preference from a plurality of contents, and a technical field of a computer program that causes a computer to function as such a list generation apparatus. .

  With the start of terrestrial digital broadcasting in 2003, network functions are becoming essentially essential functions for television broadcast receivers (hereinafter referred to as “TV terminals” as appropriate). Interactive functions and services such as information collection and information provision from users have been proposed for television terminals that have only received information from television broadcast stations using this network function. For example, in order to further develop the advertising business, many services have been proposed that collect user operation histories and present advertisement information that matches user preferences. In addition, as disclosed in Patent Documents 1 and 2, a service for selectively presenting not only advertising information but also a TV program that matches a user's preference has been proposed.

JP 2006-229998 A JP-T-2006-524209

  In the service that selectively presents the TV program that matches the user's preference described above, it is related to the user profile input by the user from the program description information included in the EPG (Electric Program Guide) information superimposed on the broadcast wave. By searching for a keyword to be performed, a television program that matches the user's preference is selectively presented. However, the user's preferences are often contradictory and may not always be determined in one direction. For example, even if the user likes a serious documentary program, he often likes to watch a light variety program. In addition, it is not always the case that one user views the television, and there are many cases where a plurality of users such as family members view one television at the same time. Furthermore, in the case of news programs and the like, a single TV program often includes a plurality of topics, and even a program that is unnecessary for the user may be determined to match the user's preference. . For this reason, the above-described method has a technical problem in that it does not always selectively present a television program that matches the preference of the user (or all of a plurality of users who watch at the same time). Yes.

  The present invention has been made in view of, for example, the above-described conventional problems. For example, a list generating apparatus and method, and a computer that can more appropriately present content that matches the user's preference that can be contradictory, can be provided. It is an object of the present invention to provide a computer program that functions as a simple list generation device.

  In order to solve the above-described problem, the list generation apparatus according to claim 1 is a first acquisition unit configured to acquire a plurality of content parameters indicating a feature of each content for each of a plurality of content feature items for each of the plurality of contents. A second acquisition unit that acquires a plurality of user parameters indicating characteristics of a user who views the plurality of contents for each of a plurality of user feature items; and the plurality of content feature items and the plurality of items for each of the plurality of contents Based on the association with the user feature item, a calculation unit that performs a predetermined calculation process using the plurality of content parameters and the plurality of user parameters, and the user based on a result of the calculation process by the calculation unit A content list in which at least some of the plurality of contents are listed in an order that matches the user's preference And a generating means for generating a.

  In order to solve the above problem, the list generation method according to claim 10 is a first acquisition step of acquiring a plurality of content parameters indicating a feature of each content for each of a plurality of content feature items for each of a plurality of contents. A second acquisition step of acquiring a plurality of user parameters indicating characteristics of a user who views the plurality of contents for each of a plurality of user feature items; and for each of the plurality of contents, the plurality of content feature items and the plurality of content parameters Based on the correspondence with the user feature item, a calculation step of performing a predetermined calculation process using the plurality of content parameters and the plurality of user parameters, and the user based on a result of the calculation process by the calculation unit A content list in which at least some of the plurality of contents are listed in an order that matches the user's preferences. And a generating step of generating and.

  In order to solve the above-described problem, a computer program according to claim 11 includes: a first acquisition unit configured to acquire a plurality of content parameters indicating a feature of each content for each of a plurality of content feature items for each of the plurality of contents; Second acquisition means for acquiring a plurality of user parameters indicating characteristics of a user who views the plurality of contents for each of a plurality of user feature items, and for each of the plurality of contents, the plurality of content feature items and the plurality of content parameters In accordance with the association with the user feature item, a calculation unit that performs a predetermined calculation process using the plurality of content parameters and the plurality of user parameters, and based on a result of the calculation process by the calculation unit, A list in which at least some of the plurality of contents are listed in the order that matches the preference. A computer program for controlling a computer provided in a list generation device comprising a generation means for generating an item list, the computer comprising: the first acquisition means; the second acquisition means; the calculation means; and the generation means It functions as at least a part of them.

  The operation and other advantages of the present invention will become apparent from the embodiments described below.

  Hereinafter, as the best mode for carrying out the invention, description will be given of embodiments relating to a list generation apparatus and method and a computer program of the present invention.

(Embodiment of List Generation Device)
An embodiment according to the list generation apparatus of the present invention is configured to obtain, for each of a plurality of contents, a first acquisition unit that acquires a plurality of content parameters indicating characteristics of each content for each of a plurality of content feature items; Second acquisition means for acquiring a plurality of user parameters indicating the characteristics of the user to be viewed for each of a plurality of user feature items, and correspondence between the plurality of content feature items and the plurality of user feature items for each of a plurality of contents The calculation means for performing a predetermined calculation process using the plurality of content parameters and the plurality of user parameters, and based on the result of the calculation process by the calculation means, in the order matching the user's preference Generation means for generating a content list in which at least a part of a plurality of contents is listed Equipped with a.

  According to the embodiment of the list generation device of the present invention, when a content list in which a plurality of contents (or a part thereof) are listed in the order that matches the user's preference is generated, first, first acquisition means and A plurality of content parameters and a plurality of user parameters are acquired by the operation of the second acquisition means. Here, it may be configured to acquire a plurality of content parameters and a plurality of user parameters that are generated in advance or recorded on a recording medium, or a plurality of content parameters and a plurality of users as appropriate at each acquisition. You may comprise so that at least one of a parameter may be newly produced | generated.

  The plurality of content parameters indicate content features for each of a plurality of content feature items including a trend feature item, a mania feature item, a family feature item, an animation feature item, a drama feature item, and the like, which will be described later. More specifically, the plurality of content parameters are parameters obtained by quantifying the content features for each of the plurality of content feature items. For example, if a numerical value of a content parameter corresponding to a trend feature item among a plurality of content parameters indicating a feature of a certain content is relatively large, it can be understood that the content is content having a strong trend element. Also, even if the value of the content parameter corresponding to the trend feature item is not as large as the content parameter value corresponding to the family feature item, for example, if the content parameter value corresponding to the family feature item is relatively large, the content may be a family that has strong trend elements. It can be seen that the elements are correspondingly strong content.

  A plurality of user parameters are set for each user feature item including a trend feature item, a mania feature item, a family feature item, an animation feature item, a drama feature item, etc., which will be described later. The characteristics of user patterns to be classified (hereinafter the same) are shown. More specifically, the plurality of user parameters are parameters obtained by digitizing user characteristics for each of a plurality of user feature items. For example, if the numerical value of the user parameter corresponding to the trend feature item is relatively large among a plurality of content parameters indicating the characteristics of a certain user, it can be understood that the user prefers content having a strong trend element. Also, even if the user parameter value corresponding to the trend feature item is not as large as the user parameter value corresponding to the family feature item, for example, if the user parameter value corresponding to the family feature item is relatively large, the user prefers content with a strong trend element, It turns out that they like content with strong family elements accordingly.

  Thereafter, a predetermined calculation process using a plurality of content parameters and a plurality of user parameters is performed for each content in accordance with the relevance between the plurality of content feature items and the plurality of user feature items. For example, a content parameter corresponding to one content feature item of a plurality of content feature items, a user parameter corresponding to one user feature item related to one content feature item of the plurality of user feature items, and However, arithmetic processing (for example, addition processing or multiplication processing) is performed.

  Based on the result of the arithmetic processing, the operation of the generation unit generates a content list in which a plurality of contents (or a part thereof) are listed in the order that matches the user's preference. For example, a content list in which a plurality of contents (or a part thereof) are listed in descending order of a numerical value (for example, a score described later) indicating the result of arithmetic processing for each content is generated.

  As described above, in the present embodiment, a content list is generated based on a plurality of content parameters indicating the characteristics of the content for each of the plurality of content feature items and a plurality of user parameters indicating the characteristics of the user for each of the plurality of user feature items. Is done. That is, a content list and a user are classified according to a single standard (in other words, a single content feature item or a single user feature item), thereby generating a unified content list or extracting content that is determined centrally. Instead, content and users are classified based on a plurality of or multifaceted criteria such as a plurality of content feature items and a plurality of user feature items.

  For example, if a numerical value increases in order of a content parameter corresponding to a trend feature item, a content parameter corresponding to a drama feature item, and a content parameter corresponding to a mania feature item among a plurality of content parameters indicating features of a certain content The content is not only a content list generated for a user who likes content with a strong trend element, but also a content list generated for a user who likes content with strong drama characteristics or a user who likes content with strong mania characteristics. Can also be included. Similarly, among the plurality of content parameters indicating the characteristics of a certain user, the numerical value increases in the order of the content parameter corresponding to the trend feature item, the content parameter corresponding to the drama feature item, and the content parameter corresponding to the mania feature item. For example, the content list generated for the user may include not only content with a strong trend element, but also content with strong drama characteristics and content with strong mania characteristics.

  As described above, according to the list generation device according to the present embodiment, even if the user's preference is a trade-off, or even if a plurality of features are included in one content, the content that matches the user's preference A list can be generated.

  According to one aspect of the embodiment of the list generating apparatus of the present invention, the plurality of content feature items and the plurality of user feature items coincide with each other, and the calculation unit includes the same feature item as the calculation process. Multiplying the plurality of content parameters and the plurality of user parameters every time and calculating a sum of multiplication results for each of the same feature items, the generation means is configured to calculate the plurality of contents in order of increasing or decreasing the sum. The content list in which at least a part of the content is listed is generated.

  According to this aspect, for example, if a plurality of content feature items include trend feature items, mania feature items, family feature items, animation feature items, and drama feature items, the plurality of user feature items also have the same content. It includes feature items (that is, trend feature items, mania feature items, family feature items, animation feature items, and drama feature items).

  Then, the computing means multiplies the content parameter corresponding to the same feature item by the user parameter and calculates the sum of the multiplication results for each feature item. For example, the content parameter corresponding to the trend feature item is multiplied by the user parameter corresponding to the trend feature item, and the sum of multiplication results for each trend feature item, mania feature item, family feature item, animation feature item, and drama feature item is summed. Is calculated.

  Thereafter, the generation unit generates a content list in which a plurality of contents are listed in order from the largest sum or the smallest.

  Thereby, as described above, even if the user's preference is a trade-off, a content list that matches the user's preference can be generated.

  In another aspect of the embodiment of the list generation apparatus of the present invention, the first acquisition means determines whether each of the plurality of contents has a feature corresponding to each of the plurality of content feature items. The plurality of content parameters are acquired based on determination information for determination.

  According to this aspect, a plurality of content parameters can be acquired relatively easily and reliably.

  In another aspect of the embodiment of the list generating apparatus of the present invention, the plurality of user parameters include, as the user characteristics, a feature of a user pattern obtained by classifying the user according to a predetermined criterion, the plurality of user feature items. Shown for each.

  According to this aspect, the plurality of users can be classified into several user groups such as a user group that prefers content with a strong trend element (in other words, a user group) and a user group that prefers content with a strong drama element. it can. For this reason, since the plurality of user parameters indicate the characteristics of the user pattern, it is possible to suppress a state in which each of the plurality of user parameters is prepared independently for each individual user. In other words, it is not necessary to prepare a plurality of individual and independent user parameters for every user. For this reason, it can suppress that a plurality of user parameters increase without limit. Furthermore, by determining which user pattern each user belongs to, a content list that matches the user's preference can be generated. For this reason, the processing load of the list generation device can be relatively reduced.

  In the aspect of the list generation device in which a plurality of user parameters indicate the characteristics of the user pattern as described above, the second acquisition means includes an input means for inputting the user characteristics, and the user based on the input characteristics. It may be configured to include a determination unit that determines which user pattern belongs to.

  With this configuration, it is relatively easy to determine which user pattern each user belongs to, so content that matches the user's preference while relatively reducing the processing load on the list generation device A list can be generated.

  As described above, in the aspect of the list generation device in which the plurality of user parameters indicate the characteristics of the user pattern, the user pattern is a group of users including a plurality of users classified according to the predetermined criterion as a predetermined criterion. You may comprise so that it may classify | categorize by.

  With this configuration, it is possible to generate a content list that matches not only a single user but also a plurality of users (for example, a family, a lover, a couple, etc.) according to user group preferences. it can.

  As described above, in the aspect of the list generation device in which the plurality of user parameters indicate the characteristics of the user pattern, the plurality of user parameters indicate the characteristics of the user pattern for each of the plurality of user characteristic items and for each time period. It may be configured.

  If comprised in this way, the content list which matched the user preference for every time slot | zone can be produced | generated in consideration that the content which a user likes may change with time slots.

  In another aspect of the embodiment of the list generating apparatus of the present invention, at least one of the plurality of content parameters and the plurality of user parameters is normalized (that is, the plurality of content parameters and the plurality of user parameters). The sum of at least one of the two becomes 1).

  According to this aspect, so-called normalization processing is performed on a plurality of content parameters and a plurality of user parameters. For this reason, it is possible to relatively easily recognize the matching ratio for each content feature item for each content. Similarly, for each user, the proportion of matching for each user feature item can be recognized relatively easily.

  Another aspect of the list generation device of the present invention further comprises a notification means for notifying the user of the content list.

  According to this aspect, the generated content list can be presented to the user, and as a result, the user can preferably select the favorite content while referring to the content list.

(Embodiment of List Generation Method)
In an embodiment according to the list generation method of the present invention, for each of a plurality of contents, a first acquisition step of acquiring a plurality of content parameters indicating characteristics of each content for each of a plurality of content feature items; A second acquisition step of acquiring a plurality of user parameters indicating the characteristics of the viewing user for each of a plurality of user feature items, and correspondence between the plurality of content feature items and the plurality of user feature items for each of the plurality of contents The calculation step of performing a predetermined calculation process using the plurality of content parameters and the plurality of user parameters, and the result of the calculation process by the calculation means, in the order that matches the user's preference Generation process for generating a content list in which at least a part of a plurality of contents is listed Equipped with a.

  According to the embodiment relating to the list generation method of the present invention, it is possible to receive the same benefits as the various benefits possessed by the above-described embodiment of the list generation apparatus of the present invention.

  Incidentally, in response to the various aspects of the embodiment of the list generation apparatus of the present invention described above, the embodiment of the list generation method of the present invention can also adopt various aspects.

(Embodiment of computer program)
According to an embodiment of the computer program of the present invention, for each of a plurality of contents, a first acquisition unit that acquires a plurality of content parameters indicating characteristics of each content for each of a plurality of content feature items, and viewing the plurality of contents Second acquisition means for acquiring a plurality of user parameters indicating user characteristics for each of a plurality of user feature items, and associating the plurality of content feature items with the plurality of user feature items for each of the plurality of contents The calculation means for performing a predetermined calculation process using the plurality of content parameters and the plurality of user parameters, and the plurality of the parameters in the order matching the user's preference based on the result of the calculation process by the calculation means. A content list that lists at least some of the content of A computer program for controlling a computer provided in a list generating device (that is, the embodiment (including various aspects) according to the above-described list generating device of the present invention), comprising: It functions as at least a part of the first acquisition means, the second acquisition means, the calculation means, and the generation means.

  According to the embodiment of the computer program of the present invention, if the computer program is read from a recording medium such as a ROM, a CD-ROM, a DVD-ROM, and a hard disk that stores the computer program and executed by the computer, or If the computer program is downloaded to a computer via communication means and then executed, the above-described embodiment of the list generating apparatus of the present invention can be realized relatively easily.

  Incidentally, in response to the various aspects of the embodiment of the list generation apparatus of the present invention described above, the embodiment of the computer program of the present invention can also adopt various aspects.

  Such an operation and other advantages of the present embodiment will be further clarified from examples described below.

  As described above, according to the embodiment of the list generation apparatus of the present invention, the first acquisition unit, the second acquisition unit, the calculation unit, and the generation unit are provided. According to the embodiment of the list generation method of the present invention, the first acquisition step, the second acquisition step, the calculation step, and the generation step are provided. According to the embodiment of the computer program of the present invention, the computer is caused to function as at least part of the first acquisition unit, the second acquisition unit, the calculation unit, and the generation unit. Accordingly, it is possible to more suitably present content that matches the user's preference that can be contradictory.

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

(1) Basic Configuration First, the program recommendation server 1 including an embodiment according to the list generation apparatus of the present invention will be described with reference to FIG. FIG. 1 is a block diagram conceptually showing the basic structure of the program recommendation server 1 according to this embodiment.

  As shown in FIG. 1, the program recommendation server 1 includes an EPG (Electric Program Guide) receiving unit 11, a program feature value calculating unit 12, a program feature value calculating parameter sheet 13, and a program information DB (Data Base) 14. A user input receiving unit 21, a user pattern analyzing unit 22, a user pattern sheet 23, a user information DB 24, a matching calculating unit 31, a recommended program list generating unit 32, and a recommended program list transmitting unit 33. .

  The EPG receiving unit 11 is configured to be able to receive EPG information superimposed on a television broadcast wave. For this reason, the EPG receiving unit 11 includes, for example, an antenna capable of receiving a broadcast wave and an extraction circuit for extracting EPG information from the received broadcast wave.

  The program feature value calculation unit 12 is based on EPG information including various information related to the television program and the program feature value calculation parameter sheet 13, and each of a plurality of television programs that can be received (or viewed) by the user. A program feature value that is a feature value (weighting value) is calculated for each of a plurality of programs. The program feature value calculated by the program feature value calculation unit 12 is recorded in the program information DB 14. The program feature value calculation operation will be described in detail later (see FIGS. 2 to 4).

  The program feature value calculation parameter sheet 13 is a data sheet (or data table) indicating a reference for calculating each program feature value of a plurality of television programs. The detailed data structure will be described later in detail (see FIG. 3).

  The program information DB 14 records the program feature value calculated by the program feature value calculation unit 12. The program information DB 14 outputs the recorded program feature value to the matching calculation unit 31 in response to a request from the matching calculation unit 31.

  The user input receiving unit 21 receives profile information related to user attributes and the like input by the user. The input of profile information by the user may be performed at a computer terminal or a television terminal connected to the program recommendation server 1 via a predetermined network such as a wired communication network or a wireless communication network. You may perform directly with respect to the said program recommendation server 1 using the keyboard etc. which are provided.

  The user pattern determination unit 22 determines which user pattern each user belongs to based on the profile information received by the user input reception unit 21. The association between the user and the user pattern is recorded in the user information DB 24 for each user, for example. The user pattern determination operation will be described later in detail (see FIG. 5).

  The user pattern sheet 23 is a data sheet (or data table) indicating user feature values that are feature values for each user pattern. The detailed data structure will be described in detail later (see FIG. 6).

  The user information DB 24 records the association between the user and the user pattern determined by the user pattern analysis unit 22. At this time, the user information DB 24 may record only the association between the user and the user pattern, or record the user feature value indicated by the user pattern in addition to the association between the user and the user pattern. Also good. Further, the user information DB 24 matches the user pattern (or the user feature value indicated by the user pattern) of the user who is the target of receiving the notification of the recommended program list in response to a request from the matching calculation unit 31. Output to the calculation unit 31.

  The matching calculation unit 31 includes a plurality of televisions that can be received by the user based on the program feature value recorded in the program information DB 14 and the association between the user and the user pattern recorded in the user information DB 24. A score that directly or indirectly indicates the priority to be recommended is calculated for each program. The score calculation operation will be described later in detail (see FIGS. 7 and 8).

  The recommended program list generation unit 32 generates a recommended program list including programs that match the user's preference based on the score for each television program calculated by the matching calculation unit 31. For example, a recommended program list in which each of a plurality of television programs that can be received by the user (or a part thereof) is arranged in descending order of score is generated.

  The recommended program list transmission unit 33 sends the recommended program list generated by the recommended program list generation unit 32 to the program recommendation server via a user (more specifically, a predetermined network such as a wired communication network or a wireless communication network). 1 to a terminal held by a user such as a television terminal or a computer terminal connected to the terminal 1.

  The EPG receiving unit 11, the program feature value calculation unit 12, the feature value calculation parameter sheet 13, and the program information DB (Data Base) 14 constitute one specific example of the “first acquisition means” in the present invention. is doing. In addition, the user input reception unit 21, the user pattern analysis unit 22, the user pattern sheet 23, and the user information DB 24 constitute one specific example of “second acquisition means” in the present invention. In particular, the user input receiving unit 21 constitutes a specific example of “input means” in the present invention, and the user pattern analysis unit 22 constitutes a specific example of “determination means” in the present invention. The matching calculation unit 31 constitutes a specific example of “calculation means” in the present invention. Further, the recommended program list generation unit 32 constitutes a specific example of “generation means” in the present invention. The recommended program list transmission unit 33 constitutes a specific example of “notification means” in the present invention. The TV program constitutes a specific example of “content” in the present invention, the program feature value constitutes a specific example of “program parameter” in the present invention, and the user feature value A specific example of “user parameter” in the invention is configured.

(2) Operation Principle Next, with reference to FIGS. 2 to 9, the operation principle of the program recommendation server 1 according to the present embodiment will be described.

(2-1) Program Feature Value Calculation Operation First, the program feature value calculation operation will be described with reference to FIG. FIG. 2 is a flowchart conceptually showing a flow of the program feature value calculation operation.

  As shown in FIG. 2, first, EPG information is received by the operation of the EPG receiving unit 11 (step S101). At this time, it is assumed that the total number of television programs that can be received by the user is N (where N is an integer of 1 or more).

  Note that in terrestrial digital broadcasting, BS digital broadcasting, and CS broadcasting currently broadcast in Japan, a user can generally obtain EPG information related to a television program broadcast from today to one week ahead. . In this case, the total number of a plurality of television programs that can be received by the user matches the total number of television programs broadcast from today to one week ahead.

  Subsequently, by the operation of the program feature value calculation unit 12, the program ID for identifying each of the plurality of television programs is i (where i is an integer equal to or greater than 1 and the initial value is 1). EPG information of the television program is read (step S102).

  Subsequently, by the operation of the program feature value calculation unit 12, the program feature ID for identifying each of the plurality of program feature items for distinguishing the television program according to the feature is j (where j is an integer of 1 or more). And an initial value is set to 1) (step S103).

  Thereafter, by the operation of the program feature value calculation unit 12, the program feature value for the program feature item of which the program feature ID is j of the television program whose program ID is i while referring to the parameter sheet 13 for calculating the program feature value Is calculated (step S104). At this time, the total number of program feature items included in the program feature value calculation parameter sheet 13 (that is, the total number of feature items for which program feature values are to be calculated) is M (where M is an integer of 1 or more). And

  Thereafter, the operation of the program feature value calculation unit 12 determines whether j is equal to or less than M, which is the total number of program feature items (step S105).

  As a result of the determination in step S105, if it is determined that j is equal to or less than M, which is the total number of program feature items (step S105: Yes), all program feature items of the TV program whose program ID is i. It is recognized that the calculation of the program feature value for is not yet completed. Therefore, in this case, after incrementing j by 1 (step S108), the operations after step S103 are repeated. That is, the program feature value for the next program feature item of the television program whose program ID is i is calculated.

  On the other hand, as a result of the determination in step S105, if it is determined that j is not less than M, which is the total number of program feature items (step S105: No), all program features of the TV program whose program ID is i. It is recognized that the calculation of the program feature value for the item has been completed. Therefore, in this case, the program feature values for all program feature items of the television program whose program ID is i are recorded in the program DB 14 (step S106).

  Subsequently, by the operation of the program feature value calculation unit 12, it is determined whether i is equal to or less than N, which is the total number of television programs (step S107).

  As a result of the determination in step S107, if it is determined that i is equal to or less than N, which is the total number of television programs (step S107: Yes), calculation of program feature values of all (that is, N) television programs. Is recognized as not complete. Accordingly, in this case, after i is incremented by 1 (step S109), the operations after step S102 are repeated. That is, the operations after step S102 are repeated until the program feature values for all the program feature items of all the television programs are calculated.

  On the other hand, as a result of the determination in step S107, if it is determined that i is not less than or equal to N, which is the total number of television programs (step S107: No), it is recognized that the calculation of the program feature values of all television programs has been completed. The Therefore, in this case, the program feature value calculation operation ends.

  Here, the operation of FIG. 2 will be described with reference to FIGS. 3 and 4 with a more specific example. FIG. 3 is a data structure diagram showing the data structure of the program feature value calculation parameter sheet 13, and FIG. 4 is a data structure diagram showing the data structure of the calculated program feature value.

  As shown in FIG. 3, the program feature value calculation parameter sheet 13 includes a keyword, a genre, a broadcast day, a broadcast station, and a broadcast time that serve as a reference when calculating a program feature value for each of a plurality of program feature items. Etc. Using such a program feature value calculation parameter sheet 13, program feature values for all program feature items are calculated for each of a plurality of television programs.

  For example, whether or not the metadata included in the EPG information of the TV program includes a detection keyword “latest, topic, information”, whether the TV program genre is an information genre, whether the TV program is broadcast The characteristics of the television program are determined by determining whether the day of the week is a holiday, whether the broadcast station of the television program is a commercial broadcast, and whether the broadcast time of the television program is from 10:00 to 12:00. Among the program feature values indicating the program feature value for the trend feature item is calculated. More specifically, for example, if the detection keyword “latest, topic, information” is included in the metadata included in the EPG information, the program feature value for the trend feature item is added. If the genre of the TV program is an information genre, the program feature value for the trend feature item is added. If the broadcast day of the television program is a holiday, the program feature value for the trend feature item is added. If the broadcast station of the television program is a commercial broadcast, the program feature value for the trend feature item is added. If the broadcast time of the television program is from 10:00 to 12:00, the program feature value for the trend feature item is added.

  Similarly, for example, whether or not the metadata included in the EPG information of the TV program includes the detection keyword “mania, deep”, whether the TV program genre is a variety genre, By determining whether the broadcast day of the week is a holiday, whether the broadcast station of the television program is a commercial broadcast, or whether the broadcast time of the television program is from 23:00 to 04:00, Of the program feature values indicating features, program feature values for mania feature items are calculated. More specifically, for example, if the detection keyword “mania, deep” is included in the metadata included in the EPG information, the program feature value for the mania feature item is added. If the genre of the TV program is a variety genre, the program feature value for the mania feature item is added. If the broadcast day of the television program is a holiday, the program feature value for the mania feature item is added. If the broadcast station of the television program is a commercial broadcast, the program feature value for the mania feature item is added. If the broadcast time of the television program is from 23:00 to 04:00, the program feature value for the mania feature item is added.

  Similarly, for example, whether or not the metadata included in the EPG information of the TV program includes a detection keyword “family, family”, whether the TV program genre is a variety genre, Depending on whether the day of the week is a holiday, whether the broadcasting station of the television program is state-run, and whether the broadcasting time of the television program is from 18:00 to 22:00, Of the program feature values shown, the program feature value for the family feature item is calculated. More specifically, for example, if the detection keyword “family, family” is included in the metadata included in the EPG information, the program feature value for the family feature item is added. If the genre of the TV program is a variety genre, the program feature value for the family feature item is added. If the broadcast day of the television program is a holiday, the program feature value for the family feature item is added. If the broadcasting station of the television program is state-run, the program feature value for the family feature item is added. If the broadcast time of the television program is from 18:00 to 22:00, the program feature value for the family feature item is added.

  Of course, it goes without saying that program feature values can be calculated in the same manner for other program feature items.

  As a result, as shown in FIG. 4, the program feature values for all the program feature items are calculated for each of the plurality of television programs. In the program information DB 14, for example, program feature values are recorded in the data structure shown in FIG.

  For example, for a TV program “Queen's Branch” with a program ID of 001, the program feature value for the trend feature item (hereinafter referred to as “trend program feature value” as appropriate) is calculated as 10, and the mania feature item is calculated. The program feature value (hereinafter referred to as “mania program feature value” as appropriate) is calculated as 3, and the program feature value for the family feature item (hereinafter referred to as “family program feature value” as appropriate) is calculated as 1. . In other words, it is understood that the TV program “Queen's Branch” with the program ID 001 is the TV program with the strongest trend feature element, the strongest mania feature element, and the next strongest family feature element. .

  Similarly, for example, for the TV program “ranking TV” with the program ID 002, the trend program feature value is calculated as 1, the mania program feature value is calculated as 10, and the family program feature value is calculated as 1. The That is, it can be seen that the TV program “ranking TV” having the program ID 002 is a TV program with relatively strong mania feature elements and relatively weak trend feature elements and family feature elements.

  Similarly, for example, for the TV program “Ababi-san” whose program ID is 003, the trend program feature value is calculated as 0, the mania program feature value is calculated as 0, and the family program feature value is calculated as 10. The That is, it can be seen that the TV program “Ababi-san” with the program ID 003 is a TV program having relatively strong family feature elements and relatively weak trend feature elements and mania feature elements.

(2-2) User Pattern Determination Operation Next, the user pattern determination operation will be described with reference to FIG. FIG. 5 is a flowchart conceptually showing a flow of the user pattern determining operation.

  As shown in FIG. 5, first, unique account information is created for each user in order to identify the user by the operation of a user account management unit (not shown) provided in the program recommendation server 1 (step S201). For example, an ID and a password are set for each user.

  Subsequently, user profile information is acquired by the operation of the user input receiving unit 21 (step S202). Specifically, after the user logs in to the program recommendation server 1 connected to the user terminal (for example, a computer terminal or a television terminal) via the network using the account information, the user's attributes (for example, gender, Age, family structure, life rhythm, lifestyle, hobby, special skill, occupation, etc.) are input as profile information. The input profile information is transmitted to the program recommendation server 1 via the network, and the transmitted profile information is received by the user input receiving unit 21.

  Subsequently, the user pattern to which the user belongs is determined based on the profile information acquired in Step S202 by the operation of the user pattern determination unit 22 (Step S203).

  Thereafter, the user pattern determined in step S203 is recorded in the user information DB 24 in a state associated with the user (step S204).

  Here, with reference to FIG. 6, a specific configuration of the user pattern sheet 23 indicating the user pattern and the user characteristic value for each user pattern will be described. FIG. 6 is a data structure diagram conceptually showing the data structure of the user pattern sheet 23. As shown in FIG.

  As shown in FIG. 6, the user pattern sheet 23 indicates user feature values for each of a plurality of user feature items for distinguishing the user pattern according to the feature for each of the plurality of user patterns. By referring to the user pattern sheet 23 and determining which user pattern each user belongs to, the user feature value of each user is uniquely determined.

  For example, for a user pattern called a trend leader whose user pattern ID for identifying a user pattern is 001, a user feature value for a trend feature item of a plurality of user feature items (hereinafter referred to as “trend user feature value” as appropriate). Is set to 10, and a user feature value for a mania feature item among a plurality of user feature items (hereinafter referred to as “mania user feature value” as appropriate) is set to 3, The user feature value (hereinafter referred to as “family user feature value” as appropriate) for the family feature item among the user feature items is set to 1.

  Similarly, for example, for a user pattern called Maniac whose user pattern ID for identifying the user pattern is 002, the trend user feature value is set to 5, and the mania user feature value is set to 10. The family user feature value is set to 1.

  Similarly, for example, for a user pattern called a family layer whose user pattern ID for identifying the user pattern is 003, the trend user feature value is set to 3, and the mania user feature value is set to 1. The family user feature value is set to 10.

  In the present embodiment, the user feature item is preferably the same as the program feature item. However, as described later, they are not necessarily the same.

(2-3) Matching Calculation Operation and Recommended Program List Generation Operation Next, the matching calculation operation and the recommended program list generation operation will be described with reference to FIG. FIG. 7 is a flowchart conceptually showing a flow of the matching calculation operation and the recommended program list generation operation.

  As shown in FIG. 7, first, a user pattern whose pattern ID is k (where k is an integer equal to or greater than 1 and the initial value is 1) from the user information DB 24 by the operation of the matching calculation unit 31. User feature value W = {trend user feature value (Wt_0), mania user feature value (Wt_1), family user feature value (Wt_2),..., User feature value (Wt_n)} is acquired (step S301). . Note that the total number of user patterns is P (where P is an integer of 1 or more).

  Subsequently, by the operation of the matching calculation unit 31, from the program information DB 14, the program feature value D = of the TV program whose program ID is i (where i is an integer equal to or greater than 1 and the initial value is 1). {Trend program feature value (Dt_0), Mania program feature value (Dt_1), Family program feature value (Dt_2),..., Program feature value (Dt_n)} are acquired (step S303). It is assumed that the total number of television programs is N (where N is an integer equal to or greater than 1).

  Thereafter, the operation of the matching calculation unit 31 calculates the score (in other words, the matching value) Mat of the TV program whose program ID is i and whose user pattern ID is k (step S305). . Specifically, multiplication processing is performed on the program feature value and user feature value for the same feature item, and the sum of the results of multiplication processing for each feature item is calculated as the score Mat. That is, if Wt_n and Dt_n (where n is an integer equal to or greater than 0) indicate a program feature value and a user feature value for the same feature item, Wt_1 × Dt_1 + Wt_2 × Dt_2 +... + Wt_n × Dt_nWt_0 × Dt_0 + Wt_1 × Dt_1 + Wt_2 × Dt_2 +... + Wt_n × Dt_n is calculated as the score Mat. More specifically, taking the program feature value shown in FIG. 4 and the user feature value shown in FIG. 6 as an example, trend program feature value × trend user feature value + mania user feature value × mania user feature value + family program feature Value × family user feature value is calculated as the score Mat.

  Thereafter, the score Mat calculated in step S305 is recorded in the program information DB 14 (step S306).

  Subsequently, whether or not i is equal to or less than N, which is the total number of television programs, is determined by the operation of the matching calculation unit 31 (step S307).

  As a result of the determination in step S307, if it is determined that i is equal to or less than N, which is the total number of television programs (step S307: Yes), all user patterns whose user pattern ID is k are targeted ( That is, it is recognized that the calculation of the score Mat for the N television programs is not completed. Accordingly, in this case, after i is incremented by 1 (step S309), the operations after step S303 are repeated. That is, the score Mat for all television programs targeting the user pattern whose user pattern ID is k is calculated.

  On the other hand, as a result of the determination in step S307, if it is determined that i is not less than or equal to N which is the total number of television programs (step S107: No), all user patterns whose user pattern ID is k are targeted. It is recognized that the calculation of the program feature value of the television program has been completed. Accordingly, in this case, it is subsequently determined whether or not k is equal to or less than P, which is the total number of user patterns (step S308).

  As a result of the determination in step S308, if it is determined that k is equal to or less than P which is the total number of user patterns (step S308: Yes), calculation of the score Mat for all user patterns has not been completed. It is recognized. Accordingly, in this case, after incrementing k by 1 (step S310), the operations after step S301 are repeated. That is, the score Mat for all user patterns is calculated.

  On the other hand, as a result of the determination in step S308, if it is determined that k is not less than or equal to P, which is the total number of user patterns (step S308: No), the scores Mat of all television programs targeting all user patterns. It is recognized that the calculation of has been completed. Therefore, in this case, the score Mat calculation operation ends.

  Here, a specific example of the score Mat will be described with reference to FIG. FIG. 8 is an explanatory diagram showing a specific example of the score Mat. It is assumed that the score Mat shown in FIG. 8 is calculated using the program feature value shown in FIG. 4 and the user feature value shown in FIG.

  As shown in FIG. 8, the score Mat of the television program “Queen's branch” with the program ID “001” targeting the user pattern “trend leader” with the user pattern ID “001” is 10 × 10 + 3 × 3 + 1 × 1. = 110. Similarly, the score Mat of the TV program “ranking TV” having the program ID 002 and targeting the user pattern of the trend leader whose user pattern ID is 001 is 1 × 10 + 10 × 3 + 1 × 1 = 41. Similarly, the score Mat of the television program “Ababi-san” whose program ID is 003 for the user pattern “trend leader” whose user pattern ID is 001 is 0 × 10 + 0 × 3 + 10 × 1 = 10.

  Similarly, the score Mat of the television program “Queen's branch” whose program ID is 001 for a user pattern of maniac whose user pattern ID is 002 is 10 × 5 + 3 × 10 + 1 × 1 = 81. Similarly, the score Mat of the TV program “ranking TV” with the program ID “002” targeting the user pattern “maniac” with the user pattern ID “002” is 1 × 5 + 10 × 10 + 1 × 1 = 106. Similarly, the score Mat of the television program “Ababi-san” with the program ID “003” targeting the user pattern “Maniac” with the user pattern ID “002” is 0 × 5 + 0 × 10 + 10 × 1 = 10.

  Similarly, the score Mat of the television program “Queen's branch” whose program ID is 001 for the user pattern of the family layer whose user pattern ID is 003 is 10 × 3 + 3 × 1 + 1 × 10 = 43. . Similarly, the score Mat of the TV program “ranking TV” having the program ID 002 and targeting the user pattern of the family layer whose user pattern ID is 003 is 1 × 3 + 10 × 1 + 1 × 10 = 23. Similarly, the score Mat of the television program “Ababi-san” whose program ID is 003 for the user pattern of the family layer whose user pattern ID is 003 is 0 × 3 + 0 × 1 + 10 × 10 = 100.

  In FIG. 7 again, the recommended program list is generated based on the score Mat calculated in step S305 (in other words, the score Mat recorded in the program information DB 14 in step S306) by the operation of the recommended program list generation unit 32. It is generated (step S311). For example, a recommended program list may be generated by arranging a plurality of TV programs in descending order of the score Mat, or, among the N TV programs, L (in the order L is the largest) A recommended program list may be generated by arranging television programs of an integer of 1 or more. However, the recommended program list is preferably generated by arranging the user patterns to which the user who is the target of receiving the recommended program list belongs, in descending order of the score Mat.

  More specifically, taking the score shown in FIG. 8 as an example, when presenting a recommended program list to users belonging to a user pattern called a trend leader, “Queen's Branch”, “Ranking TV”, “Abalone” It is preferable that a recommended program list arranged in the order of "" is generated. Further, when the recommended program list is presented to a user belonging to the user pattern of Maniac, it is preferable that a recommended program list arranged in the order of “ranking TV”, “queen's branch”, and “Ababi-san” is generated. Further, when the recommended program list is presented to users belonging to the user pattern of the family layer, it is preferable that a recommended program list arranged in the order of “Ababi-san”, “Queen's Branch”, “Ranking TV” is generated. .

  Thereafter, the recommended program list generated in step S311 is transmitted to the user by the operation of the recommended program list transmission unit 33 (step S312).

  Here, the recommended program list will be described with reference to FIG. FIG. 9 is an explanatory diagram conceptually showing the recommended program list.

  As shown in FIG. 9, a recommended program list in which priority, program title name, broadcast date and time, genre, and score are arranged in descending order is a recommended program list that is actually presented to the user. As an example. The recommended program list shown in FIG. 9 includes not only a single genre television program but also a plurality of genre television programs. In order to emphasize the ease of viewing the recommended program list for the user, the recommended program list does not necessarily have to be arranged in order of increasing score Mat. For example, the program date and time are shown together with the score Mat and its rank. TV programs may be arranged in order.

  As described above, in this embodiment, a plurality of program feature values indicating the characteristics of a television program for each of a plurality of program feature items and a plurality of user feature values indicating a feature of a user pattern for each of a plurality of user feature items. Based on this, a recommended program list is generated. That is, the above-described background art operation of generating a unified recommended program list by classifying television programs and users according to a single criterion (in other words, a single program feature item or a single user feature item). Instead, television programs and users are classified based on a plurality of or multifaceted criteria such as a plurality of program feature items and a plurality of user feature items.

  If television programs and users are classified based on a single standard, a recommended program list including only dramas tends to be generated or generated for users who prefer dramas. In other words, for users who like dramas, the fact that the user's preferences can be contradictory is not taken into account, and therefore, the recommended program list including TV programs other than dramas is not generated or is less likely to be generated. However, according to the present embodiment, as shown in FIG. 9, even if a user likes a drama, if the user also likes a TV program of another genre, other genres (for example, information A recommended program list including corresponding TV programs such as programs and movies is presented.

  Thus, according to the program recommendation server 1 according to the present embodiment, it is possible to generate a recommended program list that matches the user's preference even if the user's preference is a trade-off.

  The number of program feature items may be increased or decreased as appropriate. In this case, if a reference for calculating a program feature value for a program feature item of a new concept is added to the program feature value calculation parameter sheet 13, a program feature item of a new concept can be added. On the other hand, if the reference for calculating the program feature value for the program feature item that is no longer needed is deleted from the program feature value calculation parameter sheet 13, the program feature item that is no longer needed can be deleted. Also, as a criterion for calculating a program feature value for a program feature item, a new criterion may be added as appropriate, or an unnecessary criterion may be deleted as appropriate. Thus, according to the present embodiment, by adding or deleting items to the program feature value calculation parameter sheet 13, the program feature items can be changed or the program feature value calculation criteria for the program feature items can be changed. Can be done relatively easily. In addition, the program feature value calculation parameter sheet 13 makes it possible to centrally manage program feature items and criteria for calculating program feature values.

  Similarly, user patterns, user feature items, and user feature values can be changed relatively easily by adding or deleting items to the user pattern sheet 23. In addition, the user pattern sheet 23 can manage user patterns and user characteristic items in an integrated manner.

  Here, a modification of the user pattern will be described with reference to FIGS. 10 and 11 in relation to the change of the user pattern. Here, each of FIGS. 10 and 11 is a data structure diagram conceptually showing a modification of the user pattern.

  As illustrated in FIG. 10, not only a user pattern targeting a single user but also a user pattern targeting a plurality of users may be included. For example, the example shown in FIG. 10 includes a user pattern for mother and child. In addition, for example, user patterns targeting “friends”, “family”, “lovers”, “couples”, and the like are examples of user patterns targeting a plurality of users.

  Furthermore, when a user pattern for a plurality of users is included, a plurality of user patterns for a plurality of users may be included. For example, in the example shown in FIG. 10, a user pattern in which mother is given priority among mothers and children (in other words, the mother's television program selection right is given priority) and a child in mother and child is given priority (in other words, child TV program selection is selected). User pattern). For example, for a user pattern of mother-child_mother priority with a user pattern ID 004 for identifying the user pattern, the drama user feature value is set to 10 and the animation user feature value is set to 3, The family user feature value is set to 1. Also, for a user pattern of mother-child_child priority whose user pattern ID for identifying the user pattern is 005, the drama user feature value is set to 1, and the animation user feature value is set to 10. The family user feature value is set to 5.

  In this way, a user who views a TV program is considered by including a user pattern for a plurality of users, considering the fact that a plurality of users such as family members often view one TV at the same time. Regardless of whether the number of users is one or more, a recommended program list that matches the user's preference can be generated.

  When setting a plurality of user patterns for a plurality of users, the user patterns may be selected by the user. In the example shown in FIG. 10, either the user pattern “mother and child_mother priority” or the user pattern “mother and child_child priority” is selected by the user, and the recommended program list is generated using the selected user pattern. It may be generated. Alternatively, for example, by the operation of the program recommendation server 1, an appropriate user pattern may be automatically selected according to a time zone or a day of the week, as in the example shown in FIG. In the example shown in FIG. 10, for example, in a certain time zone (or a certain day of the week), a recommended program list is generated using a user pattern of “mother and child_mother priority”, and other time zones (or other times) The recommended program list may be generated using a user pattern of “mother and child_child priority”.

  As shown in FIG. 11, even when a single user is targeted, for example, a user pattern may be switched according to a time zone. In the example shown in FIG. 11, a user belonging to the user pattern whose user pattern ID is 006 belongs to a user pattern called a trend leader during a time period from 6:00 to 8:00, and a time from 18:00 to 22:00. The band may belong to a user pattern called a family layer, and the time zone from 22:00 to 6:00 may belong to a user pattern called business.

  Thus, for example, if it is configured to switch the user pattern according to the time zone, it matches the user's preference in consideration of the fact that what the user wants for the TV program may differ according to the time zone. A recommended program list can be generated.

  Note that once the profile information that has been input is updated, the user pattern to which the same user belongs can be changed as appropriate. For this reason, it is possible to generate a recommended program list that matches the current preference of the user. At this time, the update of the profile information may be performed by the user's own operation. Or you may comprise so that the program recommendation server 1 may update automatically, grasp | ascertaining the tendency of the television program which the user actually watched.

  Furthermore, the program feature value and the user feature value may be normalized as shown in FIGS. That is, the program feature values may be calculated so that the sum of the program feature values is 1. Similarly, the user feature values may be calculated so that the sum of the user feature values is 1. As a result, the ratio of a certain feature value to all the feature values can be grasped relatively easily. For this reason, it is possible to recommend a television program having a more specific feature or to recommend a television program that matches the user's preference having a more specific feature.

  Moreover, as shown in FIG. 14, you may comprise so that a program characteristic value may be graphed. Thereby, the feature of the television program can be grasped relatively easily. Similarly, the user feature values can be grasped relatively easily by graphing the user feature values. In this case, it is preferable that the program feature value and the user feature value are normalized.

  Furthermore, in the above-described embodiment, an example in which a plurality of program feature items and a plurality of user feature items match is described. However, the two do not necessarily need to match. In this case, it is preferable that association (or association) between a plurality of program feature items and a plurality of user feature items is determined in advance. In calculating the score, it is preferable to perform a predetermined calculation process using the program feature value for the program feature item and the user feature value for the user feature item associated with each other. For example, multiplication processing may be performed on the program feature value for the program feature item and the user feature value for the user feature item that are associated with each other, and the sum of the results of the multiplication processing may be used as the score.

  In the above-described embodiment, the program feature values of all program feature items (three in the example shown in FIG. 4) are calculated. However, some program feature items may be selectively calculated. . Specifically, by calculating program feature values for at least two or more program feature items among a plurality of program feature items, it is possible to generate a recommended program list that matches the user's preference that is contradictory. The same applies to user feature items.

  In the above-described embodiment, the program recommendation server 1 is provided separately from the television terminal owned by the user, and the recommended program list generated in the program recommendation server 1 is transmitted to the television terminal via a predetermined network. The explanation of the configuration to be advanced was advanced. This is because a function for generating a recommended program list is provided separately from the television terminal, with an emphasis on the fact that the function of the television terminal is more limited than a general-purpose computer. Since only one common server is required, the recommended program list can be generated at a relatively low cost without increasing the cost of the television terminal. However, it goes without saying that the television terminal itself owned by the user may include each component (or part thereof) included in the program recommendation server 1 as its internal configuration.

  In the above-described embodiment, an example in which a recommended program list is generated for a television program using television broadcast waves (that is, television broadcast content) has been described. However, the recommended program list may be generated not only for the TV broadcast content but also for the content received by the user via the cable TV broadcast content or the wired communication network or the wireless communication network. Needless to say. Further, even if the content is recorded on a recording medium such as a DVD or Blu-ray Disc, or the content is streamed or downloaded via the Internet or the like, it includes a plurality of programs (that is, titles). In this case, the recommended program list may be generated by performing the above-described operation.

  In addition to the television program, the above-described program recommendation server 1 may be used for distribution of advertising content such as CM. Specifically, for example, the above-described score calculation is performed using a plurality of feature values of advertisement content and a plurality of user feature values, so that advertisement content that matches user preferences is distributed. May be.

  In the above-described embodiment, as an example of content, content including video and audio such as a TV program is given, but of course video content including only video, music content or audio content not including video, Even if it is a text content or an auxiliary content (for example, metadata) that complements or explains these various types of content, the recommended content list may be generated by performing the above-described operation. .

  The present invention is not limited to the above-described embodiments, and can be appropriately changed without departing from the gist or concept of the invention that can be read from the claims and the entire specification, and the list generation accompanying such changes Apparatuses and methods, and computer programs are also included in the technical scope of the present invention.

It is a block diagram which shows notionally the basic composition of the program recommendation server which concerns on a present Example. It is a flowchart which shows notionally the flow of calculation operation | movement of a program characteristic value. It is a data structure figure which shows the data structure of the parameter sheet | seat for program characteristic value calculation. It is a data structure figure which shows the data structure of the calculated program characteristic value. It is a flowchart which shows notionally the flow of the determination operation | movement of a user pattern. It is a data structure figure which shows notionally the data structure of a user pattern sheet. It is a flowchart which shows notionally the flow of matching calculation operation | movement and the production | generation operation | movement of a recommendation program list. It is explanatory drawing which shows the specific example of a score. It is explanatory drawing which shows a recommendation program list notionally. It is a data structure figure which shows notionally the data structure of the user pattern sheet containing the user pattern which concerns on a 1st modification. It is a data structure figure which shows notionally the data structure of the user pattern sheet | seat containing the user pattern which concerns on a 2nd modification. It is a data structure figure which shows the data structure of the normalized program characteristic value. It is a data structure figure which shows notionally the data structure of the user pattern sheet | seat containing the user pattern which shows the normalized user characteristic value. It is explanatory drawing which shows the program characteristic value graphed.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Program recommendation server 11 EPG receiving part 12 Program feature value calculation part 13 Parameter sheet for program feature value calculation 14 Program information DB
21 User Input Reception Unit 22 User Pattern Determination Unit 23 User Pattern Sheet 24 User Information DB
31 Matching Calculation Unit 32 Recommended Program List Generation Unit 33 Recommended Program List Transmission Unit

Claims (11)

First acquisition means for acquiring a plurality of content parameters indicating the characteristics of each content for each of a plurality of content feature items for each of the plurality of contents;
Second acquisition means for acquiring a plurality of user parameters indicating characteristics of a user who views the plurality of contents for each of a plurality of user characteristic items;
Calculating means for performing predetermined calculation processing using the plurality of content parameters and the plurality of user parameters in accordance with the relevance between the plurality of content feature items and the plurality of user feature items for each of the plurality of contents; ,
Generating means for generating a content list that lists at least some of the plurality of contents in an order that matches the user's preference based on a result of the arithmetic processing by the arithmetic means; apparatus. The plurality of content feature items and the plurality of user feature items match,
The computing means multiplies the plurality of content parameters and the plurality of user parameters for each identical feature item as the computation processing, and calculates a sum of multiplication results for each of the same feature items,
The list generation apparatus according to claim 1, wherein the generation unit generates the content list in which at least a part of the plurality of contents is listed in order of increasing or decreasing the sum.   The first acquisition means sets the plurality of content parameters based on determination information for determining whether each of the plurality of contents has a feature corresponding to each of the plurality of content feature items. The list generation apparatus according to claim 1, wherein the list generation apparatus acquires the list.   2. The list generation according to claim 1, wherein the plurality of user parameters indicate a feature of a user pattern obtained by classifying the user according to a predetermined criterion as the feature of the user for each of the plurality of user feature items. apparatus. The second acquisition means includes
Input means for inputting the characteristics of the user;
The list generation apparatus according to claim 4, further comprising: a determining unit that determines which user pattern the user belongs to based on the input feature.   The list generation apparatus according to claim 4, wherein the user pattern is a list in which a single user is classified according to the predetermined criterion or a group of users including a plurality of users is classified according to a predetermined criterion.   The list generating apparatus according to claim 4, wherein the plurality of user parameters indicate characteristics of the user pattern for each of the plurality of user characteristic items and for each time period.   The list generation apparatus according to claim 1, wherein at least one of the plurality of content parameters and the plurality of user parameters is normalized.   The list generation apparatus according to claim 1, further comprising notification means for notifying the user of the content list. A first acquisition step of acquiring a plurality of content parameters indicating the characteristics of each content for each of a plurality of content feature items for each of the plurality of contents;
A second acquisition step of acquiring a plurality of user parameters indicating characteristics of a user who views the plurality of contents for each of a plurality of user feature items;
A calculation step of performing predetermined calculation processing using the plurality of content parameters and the plurality of user parameters in accordance with the association between the plurality of content feature items and the plurality of user feature items for each of the plurality of contents; ,
A content list comprising: a generation step of generating a content list that lists at least some of the plurality of contents in an order that matches the user's preference based on a result of the arithmetic processing by the arithmetic means. Generation method.   For each of a plurality of contents, a first acquisition unit that acquires a plurality of content parameters indicating characteristics of each content for each of a plurality of content feature items, and a feature of a user who views the plurality of contents for each of a plurality of user feature items The plurality of content parameters and the plurality of content parameters according to the association between the plurality of content feature items and the plurality of user feature items for each of the plurality of content items. Content that lists at least a part of the plurality of contents in an order that matches the user's preference based on a calculation unit that performs a predetermined calculation process using user parameters and a result of the calculation process performed by the calculation unit A controller provided in a list generation device comprising generation means for generating a list. A computer program for controlling the Yuta, the computer, the first obtaining means, said second acquisition means, a computer program for causing to function as at least one portion of said operation means and said generating means.

Images