JP5619401B2 - Content presentation system - Google Patents

Description

  The present invention relates to a content presentation system that presents content according to the preference of a group to a group composed of a plurality of users.

  Conventionally, there has been proposed a content presentation system that presents content that matches a user's content preference based on a user's profile (hereinafter referred to as a “user attribute”) and a situation such as a day of the week or a time zone. (For example, refer to Patent Document 1). In such a content presentation system, when content is evaluated by a user, first, the result is collected as learning data in a set of “user attribute, content attribute, situation attribute, reason attribute, evaluation attribute”. Next, from the collected learning data, the relationship between the user attribute, the content attribute, the situation attribute, the reason attribute, and the evaluation attribute is expressed by a Bayesian network (this is called a preference model). Is used to estimate the user's content preference. Next, based on the estimated content preference, the content estimated to meet the user's preference is determined, and this content is presented to the user.

  Here, the user attributes are “age”, “sex”, and the like. The content attributes are “genre” such as drama and variety, “channel” and the like when the content is a television program. The situation attributes are “mood”, “time zone”, and the like. The reason attribute is “because it has a sense of superiority” or the like. The evaluation attributes are “like / dislike”, “like / dislike appearance”, and the like.

JP 2006-127452 A

  By the way, it is conceivable that the content preference is estimated for one user or a group composed of a plurality of users. The content presentation system described above can estimate content preference for one target. For this reason, in the case of estimating the content preference of one user, the above-described content presentation system estimates that the content that can be slowly viewed is preferable, and estimates the content preference of a group composed of a plurality of users. It can be determined that the exciting content is preferable.

  Here, for example, when the above-mentioned evaluation attribute is expressed in five stages of “very like”, “like”, “normal”, “dislike”, “very dislike”, and a group is composed of two users consider. The first content is content that one user evaluates as “very fond” and the other user evaluates as “very dislike”, and the second content is that both of the two users have “ It is assumed that the content is evaluated as “normal”.

  Then, the above-described content presentation system has a problem of which one of the first content and the second content should be determined to be preferable for the group. In addition, when the above-described content presentation system has a priority between two users, that is, when the opinion of one user is more influential for the group than the opinion of the other user, The relationship could not be reflected in the estimation of content preference.

  As described above, in the above-described content presentation system, it may be difficult to estimate content preferences that can be satisfied by all of a plurality of users in a group composed of a plurality of users. For this reason, in the above-described content presentation system, in a group composed of a plurality of users, it may be difficult to present content that can be satisfied by all of the plurality of users.

  Accordingly, the present invention has been made in view of the above-described problems, and an object thereof is to provide a content presentation system capable of presenting content that can be satisfied by all of a plurality of users in a group composed of a plurality of users. And

( 1 ) The present invention is provided so as to be able to communicate with a user terminal (for example, equivalent to the user terminals TA, TB, and TC in FIG. 1 ) having a short-range wireless communication function, and display means (for example, the display D in FIG. 1). Display on the basis of short-range wireless communication between the communication means (for example, equivalent to the communication unit 1 in FIG. 2) provided between the communication means and the user terminal. A viewing user identification unit (for example, corresponding to the viewing user identification unit 2 in FIG. 2) for identifying a user who is viewing the main content being viewed, and all users currently viewing identified by the viewing user identification unit. Based on the content preference estimated by the preference estimation unit (e.g., corresponding to the preference estimation unit 4 in FIG. 3) and the content estimation estimated by the preference estimation unit. Main content determining means for determining a main content to be displayed by the step (e.g., corresponding to a main content determination section 5 of FIG. 3), a communication exchanged between the communication unit and the user terminal, determining result by the main content determining means And a user information update unit (for example, equivalent to the user specific information update unit 41 in FIG. 8) that updates the user specific information based on at least one of the user estimation information , the preference estimation unit, For each viewing user identified by the viewing user identification means, user-specific information (e.g., equivalent to a user X-specific usage history, evaluation history, and questionnaire information described later) and the content preference are estimated. Based on information of all users (for example, equivalent to initial personal learning data and personal preference estimation initial model described later) A personal preference estimation model generation unit (for example, corresponding to the personal preference estimation model generation unit 46 in FIG. 8) that generates a user-specific personal preference estimation model and the viewing preference generated by the personal preference estimation model generation unit A content preference estimation unit that estimates the content preference of the group based on the individual preference estimation model unique to each user and information unique to the group (e.g., equivalent to a group decision model specific to group Y described later). (For example, corresponding to the content preference estimation unit 49 in FIG. 8) is proposed.

  According to the present invention, the content presentation system is provided with communication means, viewing user identification means, preference estimation means, and main content determination means. The communication means is provided in the vicinity of the display means, and the communication means communicates with the user terminal having the short-range wireless communication function. Here, when short-range wireless communication is exchanged between the communication means and the user terminal, the user terminal is relatively close to the display means provided in the vicinity, and possesses this user terminal. This means that the user is in a range where the main content displayed by the display means can be viewed. Therefore, the viewing user identification unit identifies the user who is viewing the main content displayed by the display unit, based on the short-range wireless communication exchanged between the communication unit and the user terminal. Further, the preference estimation unit estimates the content preference of the group composed of all the users who are currently viewed and identified by the viewing user identification unit. In addition, the main content determination means determines the main content displayed by the display means based on the content preference estimated by the preference estimation means.

  For this reason, the content presentation system always recognizes the user who is viewing the main content displayed by the display means, and displays the main content that matches the content preference of the group composed of all the users who are viewing. To display. Therefore, even if the number of users who view the main content increases or decreases while the main content is displayed on the display means, it is possible to present the main content that can be understood by all of the viewing users at all times.

  Further, according to the present invention, the preference estimation means includes the personal preference estimation model generation unit that generates the user-specific personal preference estimation model and the content preference estimation unit that estimates the group content preference. Then, the personal preference estimation model generator generates user-specific information for each user being viewed, identified by the viewing user identification means, based on user-specific information and information on all users for estimating content preferences. An individual preference estimation model was generated. In addition, the content preference estimation unit estimates the content preference of the group based on the individual preference estimation model specific to each viewing user generated by the personal preference estimation model generation unit and the information specific to the group. It was.

  For this reason, the content presentation system provided with the preference estimation means combines information unique to each user constituting the group and information specific to the group into information of all users for estimating the content preference, Estimate group content preferences. Therefore, since the characteristics such as the preferences of each user constituting the group and the characteristics of the group can be reflected in the content preference estimation, the content preference that can be satisfied by all the users constituting the group. Can be estimated.

  According to the present invention, the user-specific information is updated based on at least one of the communication exchanged between the communication unit and the user terminal and the determination result by the main content determination unit in the content presentation system. User information updating means is provided. For this reason, the content of communication and the result of the determined main content can be accumulated and used for content preference estimation. For this reason, as the user terminal is operated or the content preference of the group is estimated, it is possible to improve the degree of satisfaction of all users constituting the group with respect to the main content to be presented.

( 2 ) In the content presentation system according to ( 1 ), the content preference estimation unit is configured such that the predetermined operation is performed on the user terminal, whereby the first operation signal (for example, described later) is received from the user terminal. To the extent that the personal preference estimation model specific to the user who owns the user terminal is reflected in the estimation of the content preference of the group. When a predetermined operation is performed on the user terminal, a second operation signal (for example, corresponding to information indicating that a “power DOWN” icon to be described later is selected) is received from the user terminal. The degree to which the user's own personal preference estimation model possessing the user terminal is reflected in the estimation of the group's content preference is small. We have proposed a content presentation system that is characterized by making it easy.

  According to this invention, when the first operation signal is received from the user terminal by performing a predetermined operation on the user terminal, the user terminal is possessed by the content preference estimation unit for estimating the content preference of the group. The degree of reflecting the user-specific personal preference estimation model was increased. In addition, when a second operation signal is received from the user terminal by performing a predetermined operation on the user terminal, the content preference estimation unit estimates the content preference of the group, and the individual unique to the user possessing the user terminal The degree of reflecting the preference estimation model was reduced. Therefore, each user can change the weight of his / her preference according to the operation of the user terminal, and reflect the changed weight in the estimation of the content preference of the group.

( 3 ) The present invention relates to the content presentation system according to ( 1 ) or ( 2), based on the main content displayed on the display unit and the content preference of the group estimated by the preference estimation unit. A secondary content determination unit (e.g., corresponding to the secondary content determination unit 6 in FIG. 3) that determines secondary content different from the main content displayed on the display unit; and the display unit includes the main content determination unit In addition to the main content determined by the above, a content presentation system is proposed in which the secondary content determined by the secondary content determination means is displayed.

  According to the present invention, the content presentation system is provided with secondary content determination means, and the secondary content determination means causes the main content displayed on the display means and the group content preference estimated by the preference estimation means; Based on the above, secondary content different from the main content displayed on the display means is determined. The display means displays the secondary content determined by the secondary content determination means in addition to the main content determined by the main content determination means. For this reason, it is possible to present secondary content that can be understood by all users to all users who are viewing the main content.

( 4 ) In the content presentation system according to any one of ( 1 ) to ( 3 ), the present invention provides main content displayed on the display unit for each user who is identified by the viewing user identification unit, Based on the personal preference estimation model generated by the personal preference estimation model generation unit, tertiary content determination means (for example, FIG. 3) determines tertiary content different from the main content displayed on the display means. Tertiary content determination unit 7) and tertiary content transmission means for transmitting the tertiary content determined by the tertiary content determination unit to the corresponding user terminal (for example, corresponding to the communication unit 1 in FIG. 2). And a content presentation system characterized by comprising:

  According to the present invention, the content presentation system includes the tertiary content determination unit and the tertiary content transmission unit. Then, for each user currently viewing identified by the viewing user identification unit by the tertiary content determination unit, the main content displayed on the display unit, the personal preference estimation model generated by the personal preference estimation model generation unit, Based on the above, the tertiary content different from the main content displayed on the display means is determined. In addition, the tertiary content determined by the tertiary content determination unit is transmitted to the corresponding user terminal by the tertiary content transmission unit. For this reason, it is possible to individually present tertiary content that can be satisfied by each user to each user who is viewing the main content.

( 5 ) The present invention is provided so as to be able to communicate with a user terminal (for example, equivalent to the user terminals TA, TB, and TC in FIG. 1) having a short-range wireless communication function, and display means (for example, the display D in FIG. 1) Display on the basis of short-range wireless communication between the communication means (for example, equivalent to the communication unit 1 in FIG. 2) provided between the communication means and the user terminal. A viewing user identification unit (for example, corresponding to the viewing user identification unit 2 in FIG. 2) for identifying a user who is viewing the main content being viewed, and all users currently viewing identified by the viewing user identification unit. Based on the content preference estimated by the preference estimation unit (e.g., corresponding to the preference estimation unit 4 in FIG. 3) and the content estimation estimated by the preference estimation unit. Main content determining means for determining a main content to be displayed by the step (e.g., corresponding to a main content determination section 5 of FIG. 3), a communication exchanged between the communication unit and the user terminal, determining result by the main content determining means And group information update means (for example, equivalent to the group specific information update unit 44 in FIG. 8) for updating the group specific information based on at least one of the above, the preference estimation means, Information specific to the group (e.g., equivalent to a usage history, evaluation history, and questionnaire information specific to group Y described later), and information on all groups for estimating the content preference (e.g., group learning initial data described later and Group preference estimation), based on the content preference of the group Based on the group preference estimation model generation unit (for example, equivalent to the group preference estimation model generation unit 47 in FIG. 8) that generates the estimation model and the group preference estimation model of the group generated by the group preference estimation model generation unit. And a content preference estimation unit (e.g., corresponding to the content preference estimation unit 49 in FIG. 8) that estimates the content preference of the group.

  According to the present invention, the content presentation system is provided with communication means, viewing user identification means, preference estimation means, and main content determination means. The communication means is provided in the vicinity of the display means, and the communication means communicates with the user terminal having the short-range wireless communication function. Here, when short-range wireless communication is exchanged between the communication means and the user terminal, the user terminal is relatively close to the display means provided in the vicinity, and possesses this user terminal. This means that the user is in a range where the main content displayed by the display means can be viewed. Therefore, the viewing user identification unit identifies the user who is viewing the main content displayed by the display unit, based on the short-range wireless communication exchanged between the communication unit and the user terminal. Further, the preference estimation unit estimates the content preference of the group composed of all the users who are currently viewed and identified by the viewing user identification unit. In addition, the main content determination means determines the main content displayed by the display means based on the content preference estimated by the preference estimation means.

  For this reason, the content presentation system always recognizes the user who is viewing the main content displayed by the display means, and displays the main content that matches the content preference of the group composed of all the users who are viewing. To display. Therefore, even if the number of users who view the main content increases or decreases while the main content is displayed on the display means, it is possible to present the main content that can be understood by all of the viewing users at all times.

  According to the present invention, the preference estimation means includes a group preference estimation model generation unit that generates a group content preference estimation model and a content preference estimation unit that estimates a group content preference. Then, the group preference estimation model generation unit generates a group content preference estimation model based on information unique to the group and information on all groups for estimating the content preference. Further, the content preference estimation unit estimates the group content preference based on the group preference estimation model of the group generated by the group preference estimation model generation unit.

  For this reason, a content presentation system provided with a preference estimation means estimates the content preference of a group by combining information specific to the group with information of all groups for estimating the content preference. Therefore, information specific to the group can be reflected in the estimation of the content preference. And since the characteristic of each user who comprises this group is reflected in the information peculiar to a group, the content preference which all these users can understand can be estimated with respect to all the users which comprise a group.

  According to the present invention, group-specific information is updated in the content presentation system based on at least one of the communication exchanged between the communication unit and the user terminal and the determination result by the main content determination unit. Group information update means was provided. For this reason, the content of communication and the result of the determined main content can be accumulated and used for content preference estimation. For this reason, as the user terminal is operated or the content preference of the group is estimated, it is possible to improve the degree of satisfaction of all users constituting the group with respect to the main content to be presented.

( 6 ) In the content presentation system of ( 5 ), the present invention provides the display unit with the main content displayed on the display unit and the content preference of the group estimated by the preference estimation unit. Secondary content determining means (for example, equivalent to the secondary content determining unit 6 in FIG. 3) for determining secondary content different from the displayed main content is provided, and the display means is determined by the main content determining means In addition to the main content, a content presentation system is proposed in which the secondary content determined by the secondary content determination means is displayed.

  According to the present invention, the content presentation system is provided with secondary content determination means, and the secondary content determination means causes the main content displayed on the display means and the group content preference estimated by the preference estimation means; Based on the above, secondary content different from the main content displayed on the display means is determined. The display means displays the secondary content determined by the secondary content determination means in addition to the main content determined by the main content determination means. For this reason, it is possible to present secondary content that can be understood by all users to all users who are viewing the main content.

( 7 ) In the content presentation system according to ( 5 ) or ( 6) , the present invention relates to user-specific information (for example, user X-specific use described later) for each user who is viewing, identified by the viewing user identification unit. Based on history, evaluation history, and questionnaire information) and information on all users for estimating the content preference (e.g., equivalent to personal learning initial data and personal preference estimation initial model described later) The personal preference estimation model generation means (for example, equivalent to the personal preference estimation model generation unit 46 in FIG. 8) for generating the personal preference estimation model and the viewing user identified by the viewing user identification means, the display Based on the main content displayed on the means and the personal preference estimation model generated by the personal preference estimation model generation means. Then, a tertiary content determination unit (for example, equivalent to the tertiary content determination unit 7 in FIG. 3) for determining a tertiary content different from the main content displayed on the display unit, and a determination by the tertiary content determination unit And a tertiary content transmitting means (for example, corresponding to the communication unit 1 in FIG. 2) for transmitting the tertiary content to the corresponding user terminal.

  According to the present invention, the content presentation system includes the personal preference estimation model generation means, the tertiary content determination means, and the tertiary content transmission means. Then, the personal preference estimation model generation means creates a user-specific information based on the user-specific information and the information on all users for estimating the content preference for each viewing user identified by the viewing user identification means. A personal preference estimation model was generated. In addition, for each user currently viewing identified by the viewing user identification unit by the tertiary content determination unit, the main content displayed on the display unit, the personal preference estimation model generated by the personal preference estimation model generation unit, Based on the above, the tertiary content different from the main content displayed on the display means is determined. In addition, the tertiary content determined by the tertiary content determination unit is transmitted to the corresponding user terminal by the tertiary content transmission unit. For this reason, it is possible to individually present tertiary content that can be satisfied by each user to each user who is viewing the main content.

( 8 ) In the content presentation system according to any one of (1) to ( 7 ), the present invention provides a new user (for example, a midway participation user described later) by the viewing user identification unit while the main content is displayed by the display unit. Is identified, a first digest content generation unit (for example, a digest content indicating an outline of a portion from the top of the main content being displayed until the new user is identified) (for example, Corresponding to the digest content generation unit 8 in FIG. 3) and first digest content transmission means for transmitting the digest content generated by the first digest content generation means to the user terminal of the newly identified user ( For example, equivalent to the communication unit 1 in FIG. A content presentation system is proposed.

  According to the present invention, the content presentation system is provided with the first digest content generation means and the first digest content transmission means. Then, when a new user is identified by the viewing user identification unit while the main content is being displayed by the display unit by the first digest content generation unit, a new user is identified from the top of the displayed main content. We decided to generate digest content that gives an overview of the above parts. In addition, the first digest content transmission unit transmits the digest content generated by the first digest content generation unit to the user terminal of the newly identified user. For this reason, the digest content indicating the outline of the portion that the user missed is transmitted to the user terminal of the user who has started viewing the main content displayed by the display means. Therefore, the user can view the digest content on his / her user terminal and recognize the outline of the missed portion.

( 9 ) The present invention relates to the content presentation system according to any one of (1) to ( 8 ), wherein a user who cannot be identified by the viewing user identification unit during display of the main content by the display unit (for example, a later-described midway departure) When a corresponding user) occurs, second digest content generation means (for example, a diagram) that generates digest content indicating an outline of the main content being displayed after the user cannot be identified until the end. 3 digest content generation unit 8) and second digest content transmission means (for example, the digest content generated by the second digest content generation means) to the user terminal of the user who can no longer be identified (for example, Equivalent to the communication unit 1 in FIG. We have proposed a content presentation system characterized by this.

  According to the present invention, the content presentation system is provided with the second digest content generation means and the second digest content transmission means. When the second digest content generation unit generates a user who cannot be identified by the viewing user identification unit while the main content is being displayed by the display unit, the user cannot be identified from the displayed main content until the end. For this part, digest content showing the outline is generated. Further, the digest content generated by the second digest content generation means is transmitted to the user terminal of the user who can no longer be identified by the second digest content transmission means. For this reason, the digest content which shows the outline | summary of the part which this user missed is transmitted to the user terminal of the user who stopped viewing the main content displayed by the display means. Therefore, the user can view the digest content on his / her user terminal and recognize the outline of the missed portion.

  According to the present invention, in a group composed of a plurality of users, it is possible to present content that can be satisfied by all of the plurality of users, and as a result, the satisfaction of the entire group can be improved.

It is a figure which shows schematic structure of the content presentation system which concerns on one Embodiment of this invention. It is a block diagram which shows the structure of the set top box with which the said content presentation system is provided. It is a block diagram which shows the structure of the server with which the said content presentation system is provided. It is a figure which shows the example of a display with which the said content presentation system is provided, and a user terminal. It is a figure which shows the example of a display of the said display and the said user terminal. It is a figure which shows the example of a display of the said display and the said user terminal. It is a figure which shows the example of a display of the said display and the said user terminal. It is a block diagram which shows the structure of the preference estimation part with which the said server is provided. It is a figure for demonstrating the personal preference initial model production | generation part with which the said preference estimation part is provided. It is a figure for demonstrating the group preference initial model production | generation part with which the said preference estimation part is provided. It is a figure for demonstrating the personal preference estimation model production | generation part with which the said preference estimation part is provided. It is a figure for demonstrating the group preference estimation model production | generation part with which the said preference estimation part is provided. It is a figure for demonstrating the group decision model production | generation part with which the said preference estimation part is provided.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. Note that the constituent elements in the following embodiments can be appropriately replaced with existing constituent elements, and various variations including combinations with other existing constituent elements are possible. Accordingly, the description of the following embodiments does not limit the contents of the invention described in the claims.

<Configuration of content presentation system AA>
FIG. 1 is a diagram showing a schematic configuration of a content presentation system AA according to an embodiment of the present invention. The content presentation system AA includes a user terminal TA owned by the user UA, a user terminal TB owned by the user UB, a user terminal TC owned by the user UC, and a set top provided to be able to communicate with the user terminals TA to TC. A box STB, a server S provided so as to be communicable with the set top box STB, and a display D for displaying contents are provided.

  Each of the user terminals TA to TC includes a display screen T1 that displays an image, an operation unit T2 that can be operated by the user, and a communication unit (not shown) that communicates with the set-top box STB. Each of the user terminals TA to TC transmits a unique signal for each of the user terminals TA to TC to the set top box STB by short-range wireless communication every time a predetermined time elapses. In addition, when each of the user terminals TA to TC is in a range where short-range wireless communication with the set top box STB is possible, a signal is transmitted to the set top box STB according to the operation of the operation unit T2 by the user. Receive a signal from the set top box STB.

  The display D is connected to the set top box STB, and displays a later-described main content and a later-described secondary content received from the set-top box STB on the display screen D1.

  The set top box STB is disposed in the vicinity of the display D. Here, for example, when the set top box STB receives a signal specific to the user terminal TA transmitted from the user terminal TA, the user terminal TA is relatively close to the display D in which the set top box STB is disposed in the vicinity. In other words, the user UA possessing the user terminal TA is in a range where the main content displayed on the display screen D1 of the display D can be viewed. Therefore, the set-top box STB determines a user who is viewing the main content displayed on the display screen D1 of the display D depending on whether or not a unique signal is received by the short-range wireless communication for each of the user terminals TA to TC. Identify.

  The server S determines main content and secondary content that can be satisfied by all users identified by the set-top box STB that the main content displayed on the display screen D1 of the display D is being viewed. Further, for each user identified by the set-top box STB that the main content displayed on the display screen D1 of the display D is being viewed, the server S determines the later-described tertiary content that can be satisfied by each user. . The main content and the secondary content determined by the server S are transmitted to the display D via the set top box STB. The tertiary content determined by the server S is transmitted to the user terminal of the corresponding user via the set top box STB.

  In addition, the server S generates a digest content indicating an outline of this portion of a portion missed by the user who has started or stopped viewing the main content. The digest content generated by the server S is transmitted to the user terminal of the corresponding user via the set top box STB.

<Configuration of set top box STB>
FIG. 2 is a block diagram showing a configuration of the set top box STB. The set top box STB includes a communication unit 1 and a viewing user identification unit 2.

  The communication unit 1 communicates with a communication unit (not shown) provided in each of the user terminals TA to TC. Specifically, the short-range wireless communication is possible with the set-top box STB among the signals unique to the user terminals TA to TC transmitted from the user terminals TA to TC by the short-range wireless communication every time a predetermined time elapses. What is transmitted from a user terminal in range is received by short-range wireless communication. In addition, in the user terminals TA to TC that are within a range in which short-range wireless communication with the set-top box STB is possible, when the operation unit T2 is operated by the user, a signal corresponding to the operation content is received by short-range wireless communication. To do. Also, the tertiary content and the digest content are transmitted to the corresponding ones of the user terminals TA to TC.

  The communication unit 1 communicates with the display D. Specifically, main content and secondary content to be displayed on the display screen D1 are transmitted to the display D.

  The communication unit 1 communicates with a communication unit 3 described later provided in the server S. Specifically, information indicating the user identified as being viewed by the viewing user identifying unit 2 described later is transmitted to the server S. Also, information indicating the main content determined by the server S and secondary content are received from the server S, and the main content and secondary content corresponding to the received information are transmitted to the display D. Moreover, the tertiary content determined by the server S and the digest content generated by the server S are received from the server S, and the received tertiary content and digest content are transmitted to the corresponding ones of the user terminals TA to TC. To do.

  The viewing user identification unit 2 identifies a user who is viewing the main content displayed on the display screen D1 of the display D depending on whether or not a unique signal is received for each of the user terminals TA to TC. Specifically, for example, when a signal specific to the user terminal TA is received by the communication unit 1, the viewing user identification unit 2 displays the user UA possessing the user terminal TA on the display screen D1 of the display D. Is identified as a user who is viewing the main content.

  The viewing user identification unit 2 identifies a user who has started viewing the main content displayed on the display screen D1 of the display D as a halfway participating user. Specifically, for example, when a signal unique to the user terminal TB is newly received by the communication unit 1 during a period from when the display of the main content is started until it is ended, the user UB of the user terminal TB is displayed. Identify as a halfway user.

  Also, the viewing user identification unit 2 identifies a user who has stopped viewing the main content displayed on the display screen D1 of the display D as a halfway withdrawal user. Specifically, for example, if a signal specific to the user terminal TC cannot be received by the communication unit 1 during a period from when the display of the main content is started to when it is ended, the user UC of the user terminal TC is set as a halfway user. Identify.

<Configuration of server S>
FIG. 3 is a block diagram showing the configuration of the server S. The server S includes a communication unit 3, a preference estimation unit 4, a main content determination unit 5, a secondary content determination unit 6, a tertiary content determination unit 7, and a digest content generation unit 8.

  The communication unit 3 communicates with the communication unit 1 provided in the set top box STB. Specifically, information indicating the viewing user transmitted from the communication unit 1 is received. Further, information indicating the main content determined by the later-described main content determining unit 5, secondary content determined by the later-described secondary content determining unit 6, and tertiary content determined by the later-described tertiary content determining unit 7 Or the digest content produced | generated by the digest content production | generation part 8 mentioned later is transmitted to set top box STB.

  The preference estimation unit 4 estimates the content preference of each user who is viewing the main content displayed on the display screen D1 of the display D based on the information received by the communication unit 3 indicating the viewing user. The content preference of a group composed of all users who are viewing the main content displayed on the display screen D1 of the display D is estimated. Details of the preference estimation unit 4 will be described later with reference to FIG.

  Based on the content preference of the group being viewed estimated by the preference estimation unit 4, the main content determination unit 5 determines the main content that can be convinced by all users belonging to this group, that is, all the users currently viewing, from the plurality of main contents. Decide from the inside. Here, the main content is, for example, a video stored in a main content storage unit (not shown) or a terrestrial digital broadcast video received by an antenna unit (not shown). When the main content determination unit 5 determines main content different from the main content displayed on the display screen D1 of the display D, information indicating the determined main content is transmitted to the set top box STB, and the main content corresponding to this information is transmitted. Is transmitted from the set top box STB to the display D and displayed on the display screen D1. This main content will be described in detail later with reference to FIGS.

  The secondary content determination unit 6 relates to the main content being displayed based on the main content displayed on the display screen D1 of the display D and the content preference of the group being viewed estimated by the preference estimation unit 4. In addition, secondary contents that can be convinced by all users who are viewing are determined from a plurality of secondary contents. Here, the secondary content is, for example, an advertisement or Web information. When the secondary content is determined by the secondary content determination unit 6, the determined secondary content is transmitted to the display D via the set top box STB and displayed on the display screen D1. This secondary content will be described later in detail with reference to FIGS.

  The tertiary content determination unit 7 relates to the main content being displayed based on the main content displayed on the display screen D1 of the display D and the content preference of each user being viewed estimated by the preference estimation unit 4. In addition, the tertiary content that can be convinced by the viewing user is determined from a plurality of tertiary contents for each viewing user. Here, the tertiary content is, for example, an advertisement or Web information. When the tertiary content is determined by the tertiary content determination unit 7, the determined tertiary content is transmitted to the corresponding one of the user terminals TA to TC via the set top box STB and displayed on the display screen T1. Is done. This tertiary content will be described later in detail with reference to FIGS.

  When the viewing user identification unit 2 identifies a halfway participating user or a halfway withdrawal user, the digest content generation unit 8 displays a digest content indicating an outline of a portion of the displayed main content that the halfway participation user or the halfway withdrawal user missed. Generate. Specifically, for example, when the user UB is identified as a mid-participating user by the viewing user identifying unit 2, the user UB is identified as a mid-participating user from the top of the main content displayed on the display screen D1 of the display D. The portion up to this time is summarized into one or more still images and generated as digest content. Further, for example, when the user UC is identified as a halfway withdrawal user by the viewing user identifying unit 2, the main content being displayed on the display screen D1 of the display D is identified as the halfway withdrawal user until the end. Are summarized into one or more still images and generated as digest contents. The digest content is generated using, for example, a technique disclosed in Japanese Patent Application Laid-Open No. 2009-76970. When the digest content is generated by the digest content generation unit 8, the generated digest content is transmitted to the corresponding one of the user terminals TA to TC via the set top box STB and displayed on the display screen T1. This digest content will be described in detail later with reference to FIGS.

  Hereinafter, the main content, the secondary content, the tertiary content, and the digest content will be described with reference to FIGS.

  FIG. 4 shows a state where the user UA and the user UB are viewing the main content displayed on the display screen D1 of the display D.

  In the approximate center of the display screen D1 of the display D, the main content CON1 is displayed. On the display screen D1 of the display D, above the area where the main content CON1 is displayed, an icon IA indicating that the user UA is viewing, an icon IB indicating that the user UB is viewing, Is displayed.

  In the display screen D1 of the display D, the Web information CON21A determined by the secondary content determination unit 6 is displayed to the right of the area where the main content CON1 is displayed, and below the area where the main content CON1 is displayed. The advertisement CON22A determined by the secondary content determination unit 6 is displayed. The user UA can select the Web information CON21A and the advertisement CON22A by operating the operation unit T2 of the user terminal TA, and can display information corresponding to the selected information on the display screen D1 of the display D. According to this, for example, when the Web information CON21A is a URL (Uniform Resource Locator), the Web page of the URL selected by the user UA is displayed on the display screen D1 of the display D. Will be displayed in the selected area. Similarly to the user UA, the user UB can select the Web information CON21A and the advertisement CON22A and display information corresponding to the selected item on the display screen D1 of the display D.

  The tertiary content CON3A is displayed on the display screen T1 of the user terminal TA of the user UA, and the tertiary content CON3B is displayed on the display screen T1 of the user terminal TB of the user UB. The user UA can select the tertiary content CON3A by operating the operation unit T2 of the user terminal TA and display information corresponding to the selected content on the display screen T1 of the user terminal TA. According to this, for example, when the tertiary content CON3A is a URL, the web page of the URL selected by the user UA is displayed. Similarly to the user UA, the user UB can also select the tertiary content CON3B and display information corresponding to the selected content on the display screen T1 of the user terminal TB.

  Here, for example, a case where a program introducing information on Shiodome is displayed as the main content CON1 will be described. In this case, as the Web information CON21A, a URL of a web page describing information that can be understood by both the user UA and the user UB among the information on sightseeing spots and restaurants in Shiodome is displayed. As the advertisement CON22A, advertisements that are acceptable to both the user UA and the user UB are displayed among advertisements in Shiodome sightseeing spots and restaurants.

  In addition, as the tertiary content CON3A, a URL of a web page describing information that can be understood by the user UA among information on sightseeing spots and restaurants in Shiodome is displayed. Further, as the tertiary content CON3B, a URL of a web page describing information that can be convinced by the user UB among information on sightseeing spots and restaurants in Shiodome is displayed. More specifically, for example, the content preference of the user UA estimated by the preference estimation unit 4 is content related to sake, and the content preference of the user UB estimated by the preference estimation unit 4 is content related to fashion. . In this case, as the tertiary content CON3A, the URL of the web page describing the information of the restaurant in Shiodome that can drink sake is displayed on the display screen T1 of the user terminal TA of the user UA. On the other hand, as the tertiary content CON3B, a URL of a web page describing information on a store in Shiodome that sells clothes and decorations is displayed on the display screen T1 of the user terminal TB of the user UB.

  FIG. 5 shows a state in which the user UC starts viewing the main content displayed on the display screen D1 of the display D after a predetermined time has elapsed in the state of FIG.

  Here, the main content CON1 displayed on the display screen D1 of the display D in FIG. 4 is also displayed in FIG. This is because the main content that can be understood by all the users UA to UC determined by the main content determination unit 5 is the same as the main content CON1 displayed on the display screen D1 of the display D in FIG.

  On the other hand, not only the above-described icons IA and IB but also an icon IC indicating that the user UC is viewing is displayed above the area where the main content CON1 is displayed in the display screen D1 of the display D. Yes.

  Further, the Web information CON21A and the advertisement CON22A displayed on the display screen D1 of the display D in FIG. 4 are replaced with the Web information CON21B and the advertisement CON22B, respectively. As the Web information CON21B, a URL of a web page describing information that can be understood not only by the user UA and the user UB but also by the user UC among the information on sightseeing spots and restaurants in Shiodome is displayed. Moreover, as advertisement CON22B, the advertisement which can be convinced not only by user UA and user UB but also user UC is displayed among advertisements of Shiodome tourist spots and restaurants.

  The display screen T1 of the user terminal TC of the user UC displays information indicating that digested digest content can be viewed from the beginning of the main content CON1 until the user UC starts viewing. . When the user UC selects to view the digest content, the digest content is displayed on the display screen T1 of the user terminal TC of the user UC. On the other hand, when the user UC selects not to view the digest content, the tertiary content CON3C is displayed on the display screen T1 of the user terminal TC of the user UC as shown in FIG.

  As the tertiary content CON3C, a URL of a web page describing information that can be understood by the user UC among information on sightseeing spots and restaurants in Shiodome is displayed. More specifically, for example, when the content preference of the user UC estimated by the preference estimation unit 4 is content related to a book, a web page describing information about a bookstore in Shiodome that sells the book as the tertiary content CON3C Is displayed on the display screen T1 of the user terminal TC of the user UC.

  FIG. 7 shows a state in which the user UB stops viewing the main content displayed on the display screen D1 of the display D after a predetermined time has elapsed in the state of FIG.

  Here, the main content CON1 displayed on the display screen D1 of the display D in FIG. 6 is also displayed in FIG. This is because the main content that both the users UA and UC determined by the main content determination unit 5 can agree with is the same as the main content CON1 displayed on the display screen D1 of the display D in FIG.

  On the other hand, in the display screen D1 of the display D, above the area where the main content CON1 is displayed, the icons IA and IC are displayed, and the icon IB indicating that the user UB is viewing is displayed. Absent.

  Further, the Web information CON21B and the advertisement CON22B displayed on the display screen D1 of the display D in FIG. 6 are replaced with the Web information CON21C and the advertisement CON22C, respectively. As the Web information CON21C, a URL of a web page describing information that can be understood by both the user UA and the user UC among the information on sightseeing spots and restaurants in Shiodome is displayed. As the advertisement CON22C, advertisements that are acceptable to both the user UA and the user UC are displayed among advertisements in Shiodome sightseeing spots and restaurants.

  In addition, on the display screen T1 of the user terminal TB of the user UB, when the user UB is identified again by the viewing user identification unit 2 as viewing some main content, the user UB is viewing the main content CON1. Summarized digest content is displayed for the part from the end to the end.

<Configuration of preference estimation unit 4>
FIG. 8 is a block diagram illustrating a configuration of the preference estimation unit 4. The preference estimation unit 4 includes a user specific information storage unit 40, a user specific information update unit 41, an individual preference initial model generation unit 42, a group specific information storage unit 43, a group specific information update unit 44, a group preference initial model generation unit 45, An individual preference estimation model generation unit 46, a group preference estimation model generation unit 47, a group decision model generation unit 48, and a content preference estimation unit 49 are provided.

  The user specific information storage unit 40 is a user of all user terminals that have performed short-range wireless communication with the set top box STB, that is, usage history, evaluation history, and questionnaire information of all users using the set top box STB. Is stored.

  When short-range wireless communication with the user terminal is performed by the communication unit 1, the user-specific information update unit 41 is a part of the information stored in the user-specific information storage unit 40 according to the content of the short-range wireless communication. The user terminal user who performed the range wireless communication is updated according to the content of the short range wireless communication. When the main content is determined by the main content determination unit 5, the information of the user identified by the viewing user identification unit 2 at the time when the main content is determined among the information stored in the user specific information storage unit 40. Update. When the secondary content is determined by the secondary content determination unit 6, the viewing user identification unit 2 identifies the secondary content among the information stored in the user specific information storage unit 40 when the secondary content is determined. Update the user's. Further, when the tertiary content is determined by the tertiary content determination unit 7, the information stored in the user specific information storage unit 40 is updated for the user of the user terminal whose tertiary content is determined. As a result of the update by the user specific information update unit 41, information of each user using the set top box STB is accumulated in the user specific information storage unit 40.

  The personal preference initial model generation unit 42 will be described with reference to FIG. The personal preference initial model generation unit 42 generates a personal preference estimation initial model which is a model for estimating a general user's content preference. Specifically, first, individual learning initial data is generated from the user's usage history, evaluation history, and questionnaire information stored in the user-specific information storage unit 40.

  Next, based on the generated personal learning initial data, a personal preference estimation initial model by a classifier such as a Bayesian network, a neural network, or a support vector machine is generated. More specifically, first, a positive model for predicting content attributes that are likely to be preferred from user attributes and situation attributes is generated using positive evaluation data included in the initial personal learning data, and included in the initial personal learning data. The negative model that predicts the content attribute that is likely to be disliked from the user attribute and the situation attribute is generated using the negative evaluation data. Next, a personal preference estimation initial model is generated using the positive model and the negative model.

  Returning to FIG. 8, the group specific information storage unit 43 stores all groups using the groups identified by the viewing user identification unit 2 at the time when short-range wireless communication is performed, that is, all set-top boxes STB are used. The group usage history, evaluation history, and questionnaire information are stored. Here, for example, in a home where there are three people, user P, user Q who is the wife of user P, and user R who is a child of user P and user Q, user P is user P and user Q. There are three groups: a couple group consisting of the user, a father and son group consisting of the user P and the user R, and a family group consisting of the user P, the user Q and the user R. It will be. If not only the user P but also the user Q and the user R are identified by the viewing user identification unit 2 when short-range wireless communication is performed with the user terminal of the user P, the group specific information storage unit 43 The usage history, evaluation history, and questionnaire information when the user P operates are stored as the family group usage history, evaluation history, and questionnaire information.

  When the short-range wireless communication with the user terminal is performed by the communication unit 1, the group-specific information update unit 44 includes information stored in the user-specific information storage unit 40 according to the content of the short-range wireless communication, The group consisting of the user of the user terminal that has performed the short-range wireless communication and the user who is viewing the main content together with the user is updated according to the content of the short-range wireless communication. In addition, when the main content is determined by the main content determination unit 5, all the users identified by the viewing user identification unit 2 at the time when the main content is determined out of the information stored in the user specific information storage unit 40. Update the group being updated. When the secondary content is determined by the secondary content determination unit 6, the viewing user identification unit 2 identifies the secondary content among the information stored in the user specific information storage unit 40 when the secondary content is determined. Update the group consisting of all users. Further, when the tertiary content is determined by the tertiary content determination unit 7, the information stored in the user specific information storage unit 40 is updated for the user of the user terminal whose tertiary content is determined. As a result of the update by the group specific information update unit 44, information of each group using the set top box STB is accumulated in the group specific information storage unit 43.

  The group preference initial model generation unit 45 will be described with reference to FIG. The group preference initial model generation unit 45 generates a group preference estimation initial model which is a model for estimating a general group content preference. Specifically, first, group learning initial data is generated for each target pattern of the group from the use history, the evaluation history, and the questionnaire information for each target pattern of the group stored in the group specific information storage unit 43.

  Next, for each target pattern of the group, based on the generated group learning initial model, a group preference estimation initial model by a classifier such as a Bayesian network, a neural network, or a support vector machine is generated. More specifically, first, using positive evaluation data included in the initial data of group learning, positive predicting content attributes that are likely to be favored from group attributes (number of people, presence of children under elementary school age, etc.) and situation attributes In addition to generating a model, a negative model that predicts a content attribute that is likely to be disliked from the group attribute and the situation attribute is generated using negative evaluation data included in the group learning initial data. Next, a group preference estimation initial model is generated for each target pattern of the group using the positive model and the negative model.

  The personal preference estimation model generation unit 46 will be described with reference to FIG. The personal preference estimation model generation unit 46 generates a personal preference estimation model that is a model for estimating a specific content preference for each user who operates his / her user terminal. Specifically, when generating a personal preference estimation model specific to the user X among the users who operate his / her user terminal, first, the usage history and evaluation history specific to the user X stored in the user specific information storage unit 40 And personal learning data specific to the user X is generated from the questionnaire information. Next, the personal learning data unique to the user X is multiplied by N (where N is a real number), and the personal learning data multiplied by N is combined with the initial personal learning data generated by the initial personal preference model generation unit 42. . Next, each parameter of the personal preference estimation initial model is updated based on a combination of the N-folded personal learning data and the personal learning initial data to obtain a personal preference estimation model unique to the user X. As described above, the personal preference estimation model unique to the user X is generated.

  The group preference estimation model generation unit 47 will be described with reference to FIG. The group preference estimation model generation unit 47 generates a group preference estimation model, which is a first model for estimating a specific content preference for each group. Specifically, when generating a group preference estimation model specific to group Y among a plurality of groups to which a user operating his / her user terminal belongs, first, the group Y specific stored in the group specific information storage unit 43 Group learning data is generated from usage history, evaluation history, and questionnaire information. Next, the group learning data is multiplied by M (where M is a real number), and the group learning data multiplied by M is combined with the group learning initial data generated by the group preference initial model generation unit 45. Next, each parameter of the group preference estimation initial model is updated based on a combination of the group learning data multiplied by M and the group learning initial data to obtain a group preference estimation initial model unique to group Y. As described above, the group preference estimation model unique to the group Y is generated.

  The group decision model generation unit 48 will be described with reference to FIG. The group decision model generation unit 48 generates, for each group, a group decision model that is a second model for estimating unique content preferences. Specifically, when a group decision model unique to group Y is generated, first, one is selected from a plurality of predetermined models such as an individual combination determination model and a priority determination model described later. Next, each parameter of the selected model is updated based on the individual preference estimation model specific to each user belonging to the group Y generated by the personal preference estimation model generation unit 46 and the information specific to the group Y. , Group Y specific group decision model. Thus, a group decision model specific to group Y is generated.

Note that the information unique to the group Y is, for example, a numerical value β _i corresponding to the priority order of the user i described later when the priority order determination model is selected.

  Returning to FIG. 8, the content preference estimation unit 49 estimates the content preference unique to the user X using the personal preference estimation model of the user X generated by the personal preference estimation model generation unit 46.

  The content preference estimation unit 49 includes a group preference estimation model specific to the group Y generated by the group preference estimation model generation unit 47 and a group decision model specific to the group Y generated by the group decision determination model generation unit 48. Either one is used to estimate the content preference specific to the group Y.

  Whether to use the group preference estimation model or the group decision making model is determined according to the number of feedbacks of the group Y with respect to the content preference estimation result. Specifically, when the number of times that the group Y has evaluated the content preference estimation result is less than a predetermined number of times, the number of times that the group Y has evaluated the content preference estimation result is determined using a group decision model. If the number of times is greater than or equal to the predetermined number, a group preference estimation model is used.

  When the group preference estimation model is used, the content preference estimation unit 49 estimates the content preference specific to the group Y based on the group preference estimation model specific to the group Y generated by the group preference estimation model generation unit 47.

On the other hand, when using the group decision-making model, the content preference estimation unit 49 first assigns the group Y based on the individual preference estimation model of each user belonging to the group Y generated by the personal preference estimation model generation unit 46. For each user belonging, the evaluation value x _i of the user i for the content x is calculated. Then, the estimated evaluation value x _i of the user i, and group Y unique group decision model generated by the group decision model generating unit 48, based on, the group Y-specific content preferences for the calculated content x To do.

  When the group decision making model is used, the weight of each user's preference is changed according to the user's operation of the user terminal, and the changed weight is reflected in the estimation of the content preference unique to the group Y.

Specifically, for example, in the state illustrated in FIG. 5, each of the user terminals TA to TC has a “power up” icon that increases the degree of reflecting its own preference in determining the main content, and its own in determining the main content. A “power DOWN” icon for reducing the degree of reflecting the preference is displayed on the display screen T1. For example, when the user UA selects the “power UP” icon, the user terminal TA transmits information indicating that the “power UP” icon is selected to the server S via the set top box STB. . Then, the server S, the content preference estimation unit 49, the evaluation value x _UA user UA for content x, multiplied by the coefficient larger than 1, which predetermined group Y-specific content preferences by using the multiplication result presume. According to this, the degree to which the preference of the user UA is reflected in the content preference unique to the group Y estimated by the content preference estimation unit 49 increases.

For example, when the user UB selects the “power DOWN” icon, the user terminal TB transmits information indicating that the “power DOWN” icon has been selected to the server S via the set top box STB. To do. Then, the server S multiplies the evaluation value x _UB of the user UB for the content x by a predetermined coefficient smaller than 1 by the content preference estimation unit 49, and uses this multiplication result to determine the content preference unique to the group Y. presume. According to this, the degree to which the preference of the user UB is reflected in the content preference unique to the group Y estimated by the content preference estimation unit 49 becomes small.

<Individual combined decision model>
The individual summation determination model is a model for selecting content having a high total evaluation value of each user belonging to the group Y. This model can be expressed as in equation (1), where F (x _i ) is the recommendation level of group Y of content x with respect to group Y.

<Priority determination model>
The priority determination model is a model in which, when there is a priority among the users belonging to the group Y, this priority can be reflected in the selection of contents, and according to the evaluation value of each user belonging to the group Y and the priority This is a model for selecting content with a high sum of the product of the number obtained by multiplying the value. This model can be expressed as Equation (2), where F (β _i , x _i ) is the recommendation level of content x for group Y.

In Expression (2), β _i indicates a numerical value corresponding to the priority order of the user i, and the numerical value increases as the priority order increases.

  According to the above set top box STB, the following effects can be obtained by providing the above-described preference estimation unit 4.

  When estimating the group's content preference using the group decision model generated based on the personal total decision model, calculate the evaluation value for the content of each user making up the group and reflect the calculation result in the estimation of the content preference it can. For this reason, since the characteristic of each user who comprises a group can be reflected in estimation of a content preference, the content preference which all the users which comprise a group can be convinced can be estimated.

  Moreover, when estimating the group content preference using the group decision model generated based on the priority determination model, not only the evaluation value for the content of each user constituting the group, but also each user in the group Priorities can also be reflected in the estimation of content preferences. For this reason, compared with the case where the content preference of a group is estimated using the group decision determination model produced | generated based on the individual total decision model, the characteristic of each user who comprises a group can be further reflected in estimation of a content preference.

  Further, when the group preference estimation is estimated using the group preference estimation model, the use history, the evaluation history, and the questionnaire information specific to the group Y can be reflected in the estimation of the content preference. For this reason, the content preference which all the users who comprise a group can be convinced can be estimated.

  Also, which one of the group preference estimation model and the group decision determination model is to be used is determined according to the number of feedbacks of the group Y with respect to the content preference estimation result. For this reason, when the number of feedbacks is less than the predetermined number, the group decision model is used on the assumption that the use history, evaluation history, and questionnaire information specific to the group Y are not sufficiently accumulated. On the other hand, when the number of feedbacks is equal to or greater than the predetermined number, the group preference estimation model is used on the assumption that the use history, the evaluation history, and the questionnaire information specific to the group Y are sufficiently accumulated. According to the above, the group preference estimation model and the group decision-making model can be properly used according to the situation, so that the content preference that can be further understood by all users constituting the group can be estimated.

  When the set-top box STB is used, the user-specific information update unit 41 stores in the user-specific information storage unit 40 based on the usage history, evaluation history, and questionnaire information of each user using the set-top box STB. Update stored information. When the set-top box STB is used, the group-specific information update unit 44 stores it in the group-specific information storage unit 43 based on the group usage history, evaluation history, and questionnaire information using the set-top box STB. Update the information that is being provided. According to the above, whenever the set top box STB is used, the information stored in the user specific information storage unit 40 and the group specific information storage unit 43 can be updated. Moreover, the content preference of a group is estimated using the information memorize | stored in these user specific information storage part 40 and group specific information storage part 43. FIG. For this reason, it is possible to more strongly reflect the information that is specific to the users constituting the group and the group in the estimation of the content preference, and the satisfaction of all the users constituting the group can be improved.

  Moreover, according to the above content presentation system AA, the following effects can be produced.

  The viewing user identification unit 2 identifies a user who is viewing the main content displayed on the display screen D1 of the display D. Then, the main content determination unit 5 determines the main content based on the content preference of the group composed of all users identified by the viewing user identification unit 2. Further, the secondary content determination unit 6 determines the secondary content based on the content preference of the group composed of all users identified by the viewing user identification unit 2. For this reason, while the main content is displayed on the display screen D1 of the display D, the user who is viewing the main content displayed on the display screen D1 of the display D even if the number of users viewing the main content increases or decreases. Main content and secondary content that can be satisfied by all these users can be presented to all the users.

  Further, the viewing user identifying unit 2 identifies a user who has started viewing the main content displayed on the display screen D1 of the display D as a midway participating user. When the viewing user identification unit 2 identifies the midway participating user by the digest content generating unit 8, the digest content generating unit 8 generates digest content indicating an outline of the portion of the displayed main content that the midway participating user missed. And the produced | generated digest content is transmitted to the user terminal of a midway participation user by the communication part 1. FIG. For this reason, the midway participating user can view the digest content on his / her user terminal and recognize the outline of the missed portion.

  Further, the viewing user identifying unit 2 identifies a user who has stopped viewing the main content displayed on the display screen D1 of the display D as a halfway leaving user. Then, when the viewing user identifying unit 2 identifies the halfway leaving user, the digest content creating unit 8 creates a digest content indicating an outline of the portion of the displayed main content that the halfway leaving user missed. Then, when the viewing user identifying unit 2 recognizes that the halfway leaving user is again viewing some main content, the communication unit 1 transmits the generated digest content to the user terminal of the halfway leaving user. For this reason, the halfway user can recognize the outline of the missed part by viewing the digest content on his / her user terminal.

  The processing of the set top box STB and the server S of the present invention is stored in a computer-readable recording medium, and the program recorded on the recording medium is read and executed by the set top box STB and the server S. The invention can be realized.

  Further, the above-described program may be transmitted from the set-top box STB and the server S storing the program in a storage device or the like to another computer system via a transmission medium or by a transmission wave in the transmission medium. . Here, the “transmission medium” for transmitting the program refers to a medium having a function of transmitting information, such as a network (communication network) such as the Internet or a communication line (communication line) such as a telephone line.

  Further, the above-described program may be for realizing a part of the above-described function. Furthermore, what can implement | achieve the above-mentioned function in combination with the program already recorded on the set top box STB and the server S, what is called a difference file (difference program) may be sufficient.

  The embodiment of the present invention has been described in detail with reference to the drawings. However, the specific configuration is not limited to this embodiment, and includes a design that does not depart from the gist of the present invention.

  For example, in the above-described embodiment, when the main content determination unit 5 determines a main content different from the main content displayed on the display screen D1 of the display D, the determined main content is displayed on the display screen D1 of the display D. However, this is not a limitation. For example, when the main content determination unit 5 determines a main content different from the main content displayed on the display screen D1 of the display D, a video indicating that viewing of the new main content is recommended is displayed on the display screen D1 of the display D. May be. According to this, the user who has viewed the main content displayed on the display screen D1 of the display D can select whether to view the main content that has been viewed up to now or to view new main content.

  Also, for example, in the above-described embodiment, the digest content is transmitted to the user terminal of the halfway exit user when the viewing user identifying unit 2 identifies that the halfway withdrawal user is again viewing some main content. However, the present invention is not limited to this. For example, when the viewing user identifying unit 2 is identified as a halfway leaving user, a digest content for the halfway leaving user may be created and transmitted to the user terminal of the halfway leaving user using a mobile phone network.

  In the above-described embodiment, for example, when the viewing user identification unit 2 identifies a minor user as a user who is viewing the main content displayed on the display screen D1 of the display D, the main content determination unit The 5 and secondary content determination unit 6 and the tertiary content determination unit 7 may not determine adult content.

  Further, for example, in the above-described embodiment, each of the user terminals TA to TC transmits a unique signal for each user terminal TA to TC to the set top box STB by short-range wireless communication every time a predetermined time elapses. The set-top box STB identifies a user who is viewing the main content displayed on the display screen D1 of the display D depending on whether or not a unique signal is received by the short-range wireless communication for each of the user terminals TA to TC. However, this is not a limitation. For example, the set top box STB outputs a search signal by short-range wireless communication, and the user terminal that has received the search signal transmits a response signal to the set top box STB. The user who is viewing the main content displayed on the display screen D1 of the display D may be identified by identifying whether the response signal is received. Alternatively, each time the predetermined time elapses, a unique signal is transmitted for each user terminal TA to TC from the user terminals TA to TC by short-range wireless communication, and the search signal is set top by short-range wireless communication. A method of outputting from the box STB may be combined to identify a user who is viewing the main content displayed on the display screen D1 of the display D.

  For example, in the above-described embodiment, the digest content is configured by one or more still images. However, the digest content is not limited thereto, and may be configured by, for example, a moving image.

  Further, for example, in the above-described embodiment, all of the main content determination unit 5, the secondary content determination unit 6, the tertiary content determination unit 7, and the digest content generation unit 8 are provided in the server S, but at least one of these is provided. May be provided in the set-top box STB. In the above-described embodiment, the personal preference estimation model generation unit 46, the group preference estimation model generation unit 47, the group decision model generation unit 48, and the content preference estimation unit 49 are all provided in the server S. At least a part of them may be provided in the set top box STB.

DESCRIPTION OF SYMBOLS 1 ... Communication part 2 ... Viewing user identification part 3 ... Communication part 4 ... Preference estimation part 5 ... Main content determination part 6 ... Secondary content determination part 7 ... Tertiary content Determination unit 8 ... digest content determination unit 40 ... user specific information storage unit 41 ... user specific information update unit 42 ... personal preference initial model generation unit 43 ... group specific information storage unit 44 Group unique information update unit 45 ... initial group preference model generation unit 46 ... individual preference estimation model generation unit 47 ... group preference estimation model generation unit 48 ... group decision model generation unit 49 ... Content preference estimation unit AA ... Content presentation system D ... Display S ... Server STB ... Set top box TA, TB, TC ... U The terminal

Claims (5)

A communication means provided to be able to communicate with a user terminal having a short-range wireless communication function, and provided in the vicinity of the display means;
Viewing user identification means for identifying a user who is viewing the main content displayed by the display means based on short-range wireless communication exchanged between the communication means and the user terminal;
Preference estimation means for estimating the content preference of a group composed of all viewing users identified by the viewing user identification means;
Main content determination means for determining the main content displayed by the display means based on the content preference estimated by the preference estimation means;
Group information update means for updating the information specific to the group based on at least one of the communication exchanged between the communication means and the user terminal and the determination result by the main content determination means;
With
The preference estimation means includes
A content preference estimation model generation unit that generates a content preference estimation model of the group based on information unique to the group and information of all groups for estimating the content preference;
A content preference estimation unit that estimates the content preference of the group based on the content preference estimation model of the group generated by the content preference estimation model generation unit;
Equipped with a,
The content preference estimation model generation unit
A group preference estimation model generating unit for generating a group preference estimation model as the content preference estimation model;
As the content preference estimation model, a group decision model generation unit that generates a group decision model,
With
The group preference estimation model generation unit
Feedback information obtained from the group for the content preference estimation result of the group;
Feedback information obtained from the group for the content preference estimation result for each group having the same target pattern as the group,
The group preference estimation model is generated using
The group decision making model generation unit
A calculation formula for obtaining a recommendation degree of content for the group, and parameters of the calculation formula;
For each user belonging to the group, feedback information obtained from the user with respect to a content preference estimation result;
Feedback information obtained from the user for the estimation result of the content preference for each user who has viewed the main content displayed by the display means;
To generate the group decision model using
The content preference estimation unit
If the number of feedbacks obtained from the group is less than the threshold for the result of estimating the content preference of the group, based on the group decision model of the group, the content preference of the group is estimated,
When the number of feedbacks obtained from the group with respect to the result of estimating the content preference of the group is equal to or greater than the threshold, the content preference of the group is estimated based on the group preference estimation model of the group. A content presentation system characterized by that. The content presentation system according to claim 1 ,
A secondary content that determines secondary content different from the main content displayed on the display means based on the main content displayed on the display means and the content preference of the group estimated by the preference estimation means Content determination means,
The content display system, wherein the display unit displays the secondary content determined by the secondary content determination unit in addition to the main content determined by the main content determination unit. The content presentation system according to claim 1 or 2 ,
An individual that generates a user-specific personal preference estimation model based on user-specific information and information on all users for estimating the content preference for each user who is being viewed, identified by the viewing user identification means Preference estimation model generation means;
The display based on the main content displayed on the display means and the personal preference estimation model generated by the personal preference estimation model generation means for each viewing user identified by the viewing user identification means Tertiary content determining means for determining tertiary content different from the main content displayed on the means;
Tertiary content transmission means for transmitting the tertiary content determined by the tertiary content determination means to a corresponding user terminal;
A content presentation system comprising: The content presentation system according to any one of claims 1 to 3 ,
When a new user is identified by the viewing user identification unit during the display of the main content by the display unit, an outline of a portion from the top of the main content being displayed until the new user is identified is displayed. First digest content generating means for generating a digest content to be shown;
First digest content transmission means for transmitting the digest content generated by the first digest content generation means to the user terminal of the newly identified user;
A content presentation system comprising: The content presentation system according to any one of claims 1 to 4 ,
When a user who can no longer be identified by the viewing user identification unit occurs during display of the main content by the display unit, an outline of the portion of the main content being displayed until the end after the user cannot be identified is shown. Second digest content generation means for generating digest content;
Second digest content transmitting means for transmitting the digest content generated by the second digest content generating means to the user terminal of the user who can no longer be identified;
A content presentation system comprising:

Images