JP2012222585A - Content recommendation system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To improve individual user satisfaction to a recommended content by simultaneously displaying recommended contents for both an individual user and combination of users.SOLUTION: By the simultaneous presentation of recommended contents for a plurality of users and for each individual user, the selection of a recommended content not influenced by tastes of other users is made easier, to prevent reduced satisfaction to the recommended content. Also, by the change of a method of displaying the recommended contents according to the number and the weighting of viewing users, the selection of the recommended content is made easier, to prevent reduced satisfaction to the recommended content. Further, by the change of the method of displaying the recommended contents according to whether an unregistered user is viewing, privacy of a registered user is protected.

Description

  The present invention relates to a content recommendation device for displaying content such as a television.

  In Patent Document 1, “Providing a television receiver capable of recommending a program of content that seems to be appropriate according to the correlation of a plurality of viewers” (see Reference Patent Document 1, paragraph 0004) is an issue. As a means for solving the problem, “a correlation storage unit for storing correlation data indicating a correlation between viewers, a viewer detection unit for detecting viewers who are watching, and a correlation between the viewers who are watching” A correlation detection unit obtained from correlation data from the correlation storage unit, a program evaluation unit that evaluates a program in accordance with the correlation detected by the correlation detection unit and calculates an evaluation value, and the program evaluation unit And a presentation means for presenting a program list having a high evaluation value to a viewer ”(see Reference Patent Document 1, paragraph 0005).

JP 2008-219182 A

  However, in Patent Document 1, although the optimum content can be displayed for the entire user, the recommended content for individual users and combinations of arbitrary users cannot be displayed at the same time.

  In addition, since the recommended content for the entire user is affected by the tastes of other users, there is also a problem that the satisfaction level of each user with respect to the recommended content decreases.

  The above problem is solved by the configuration described in the claims as an example.

  According to the present invention, the recommended content for the entire user, the recommended content for individual users, and the recommended content for any user combination are always recommended regardless of the user configuration. An apparatus that provides and displays efficiently can be provided.

It is a figure which shows a user and a content recommendation apparatus. It is a block diagram of the configuration of the content recommendation device. It is a functional block diagram of a content recommendation apparatus. It is a flowchart which shows the recommended content provision processing procedure in the content recommendation apparatus of this invention. It is a figure which shows the example of a display of recommended content. It is a flowchart which shows the recommended content provision processing procedure in the content recommendation apparatus of this invention. It is a figure which shows the example of a display of recommended content. It is a figure which shows the example of a display of recommended content. It is a functional block diagram of a content recommendation apparatus. It is a flowchart which shows the recommended content provision processing procedure in the content recommendation apparatus of this invention. It is a figure which shows the example of a display of recommended content. It is a figure which shows the example of a display of recommended content. It is a functional block diagram of a content recommendation apparatus. It is a flowchart which shows the recommended content provision processing procedure in the content recommendation apparatus of this invention. It is a flowchart which shows the recommended content provision processing procedure in the content recommendation apparatus of this invention. It is a figure which shows the example of a display of recommended content.

  An embodiment of a content recommendation device to which the present invention is applied will be described with reference to FIGS.

  FIG. 1 is a diagram showing the relationship between a user who operates and views in front of a content recommendation device and the content recommendation device, FIG. 2 is a block diagram of the configuration of the content recommendation device of the present invention, and FIG. 3 is a function of the present invention. FIG. 4 is a flowchart showing a recommended content provision processing procedure in the content recommendation device of this embodiment, and FIG. 5 is a diagram showing a display example of content displayed on the content recommendation device.

  The contents in this embodiment are a program received from a broadcasting station, a video recorded on a storage medium such as an internal or external hard disk, a video recorded on a removable medium such as a DVD or Blu-ray, and received and reproduced via a network. Mainly available video, other video, images, and audio that can be played on the display device are also included.

  The recommended content in the present embodiment refers to user attribute information such as the user's age and sex, preference information such as favorite entertainers and favorite genres, and viewing environment information such as viewing time and viewing time. The information that characterizes the user is parameterized, and based on the parameter, it is determined that it matches the user's preference and situation, and indicates the selected content.

  In FIG. 1, 101 and 102 are users who view content, and the user 102 displays recommended content using the remote controller 103. A user information acquisition unit 104 is installed in the content recommendation device 105 and acquires user information in front of the content recommendation device 105. Specifically, it can be realized by an imaging device such as a camera. Further, when a button for identifying an individual is assigned to the remote controller 103, the receiving device of the remote controller also corresponds to the user information acquisition unit 104. It is also possible to specify an individual by voice, and in this case, a voice input device such as a microphone corresponds to the user information acquisition unit 104. An information display unit 106 displays recommended content on the content recommendation device 105.

  In FIG. 5, reference numerals 501, 502, and 503 are display areas for recommended content displayed on the information display unit 106.

  In FIG. 2, reference numeral 201 denotes a user information registration unit, which is a storage area in which parameters characterizing the user such as the user attribute information, the preference information, and the viewing environment information are stored. Reference numeral 202 denotes a CPU, which executes processing in functional blocks described later. Reference numeral 203 denotes a content information storage unit that stores where content to be displayed can be acquired as recommended content and parameters that characterize the content (title, performer, genre, etc.). In the case where the content is a recorded program, which storage medium is stored in which part, and in the case of data over a network, location information such as a URL (Uniform Resource Locator) is stored.

  FIG. 3 shows functional blocks processed using the apparatus of FIG. Reference numeral 301 denotes a user identifying unit that identifies a user who is viewing in front of the content recommendation device 105. A recommended content display unit 302 displays recommended content on the information display unit 106. 303 is a recommended content calculation unit for one user that calculates recommended content that matches the preference of one user, and 304 is a recommended content for user combination that calculates recommended content that matches the preference of two users or all users. It is a calculation part.

  The user identification unit 301 identifies who is viewing in front of the content recommendation device 105 based on the information input from the user information acquisition unit 104 and the information registered in the user information registration unit 201. To do. In this embodiment, a camera is used for the user information acquisition unit 104. The camera periodically acquires images, and from the images, the user identification unit 301 calculates the feature amounts of the two face images of the user A and the user B, and the user face images registered in the user information registration unit 201 in advance. It is determined that the user with the highest matching rate is in front of the content recommendation device 105. For each user detected by the camera, the user specifying process by comparing the feature amounts is repeated, and the user in front of the content recommendation device 105 is specified. By specifying users, the number of viewing users can also be specified at the same time. The recommended content display unit 302 displays the recommended content on the content recommendation device 105. A one-user recommended content calculation unit 303 calculates recommended content for each user specified by the user specifying unit 301. As an example, by correlating a parameter registered in the user information registration unit 201 that characterizes one user with a parameter stored in the content information storage unit 203 that characterizes the content, a highly correlated content is directed to that user. A method of selecting one or a plurality of recommended contents may be considered. In this embodiment, recommended content for the user 101 and recommended content for the user 102 are calculated. The recommended content calculation unit 304 for user combinations calculates recommended content for each pattern that covers all combinations for selecting two users from all users, and recommended content for all users. Specifically, by correlating the parameters of the two users registered in the user information registration unit 201 and the parameters stored in the content information storage unit 203, content having a strong correlation is selected as recommended content. The recommended content to be selected may be one having the highest correlation or a method of selecting a plurality of recommended contents in the order of highest correlation. When only one item is selected and presented, there is little information to present, and the user operation becomes easy. When a plurality of presentations are presented, the viewing content can be selected by the user, the user's preference can be obtained from the selection result, and the content more suitable for the user's preference can be recommended.

  Next, a method for displaying recommended content when there are two users will be described with reference to FIG.

In FIG. 4, the user information acquisition unit 104 is used to acquire user information for viewing content in front of the content recommendation device 105. (Step 401)
The user identifying unit 301 identifies who the user is based on the information acquired in step 401 and the information registered in the user information registering unit 201, and calculates the number of users. (Step 402)
The recommended content calculation unit 303 for one user calculates the recommended content for each user from the data in the content information storage unit 203 for each user specified in step 402. (Step 403)
Subsequent processing is changed depending on whether the number of persons calculated in step 402 is one or two. When the number of people is one, there is no need to calculate recommended content as a combination of users. (Step 404)
When the number of users is two in step 404, the recommended content calculation unit 304 for user combination calculates recommended content that satisfies the preferences of the two users 101 and 102 at the same time. (Step 405)
When there are two users, the recommended content calculated in steps 403 and 405 is displayed on the content recommendation device 105 at the same time. (Step 406)
Since it is not necessary to display the recommended content as a combination when there is one user, the recommended content for the user specified in step 402 is displayed on the content recommendation device 105. (Step 407)
In FIG. 5, display areas 501, 502, and 503 indicate display areas and display examples of recommended content displayed on the information display unit 106. Recommended content for the user 101 is displayed in the recommended content display area 501, recommended content for the user 102 is displayed in the recommended content display area 502, and recommended content for the two users 101 and 102 is displayed in the recommended content display area 503. The uppermost part of each area is the program that best suits the user's preference, and the degree of adaptation decreases as it goes down. In this example, since user A is characterized as interested in sports programs and buildings, programs such as baseball, soccer, and world heritage are highly recommended. Since user B is characterized as interested in soccer, cooking, and buildings, programs such as soccer, cooking programs, and world heritage are recommended to the top. In the display area 503, as a recommended content that satisfies the tastes of both the user A and the user B at the same time, a travel program in which soccer, sports news, world heritage, cuisine and buildings can be enjoyed simultaneously is displayed.

  In the case of the present embodiment in which a camera is used as the user information acquisition unit 104, the user's positional relationship can be acquired from the acquired image. In the case of the present embodiment, based on this positional relationship, recommended content for user A is displayed on the left side of content recommendation device 105 on user A side, and recommended content for user B is displayed on the right side of content recommendation device 105 on user B side. By providing the recommended content display unit 302 with the function to perform, the convenience to the user can be further improved.

  As described above, in this embodiment, when a plurality of users are detected, recommended content that suits each user's preference is selected and presented, and two users are selected from all users to meet both preferences. Recommended content and recommended content that matches the preferences of all users are selected and presented. As a result, it is possible to realize both content that matches the preferences of individual users and presentation of content that matches the common preferences. Further, by displaying as shown in FIG. 5, it is possible to realize a display that is easy for the user to understand and to easily select content.

  This embodiment is a content recommendation apparatus and method that provides processing when the number of users is different from that of the first embodiment. The configuration block diagram and functional block diagram in the present embodiment are the same as those in the first embodiment, and will be described with reference to FIGS. 2 and 3 and FIGS.

In FIG.
As a result of steps 601 and 602, the subsequent processing is changed depending on whether or not there are three or more users. (Step 602)
The three users are user A, user B, and user C, respectively. Arbitrary user combinations are user AB, user AC, user BC, and user ABC. When a recommended program is calculated for each, seven recommended content display areas are required together with recommended programs for individual users. As the number of people increases, the number of combinations increases, and the content cannot be displayed in the limited display area of the content recommendation device 105, or even if it can be displayed, it is difficult for the user to see. Therefore, in this embodiment, when the number of users increases to three or more, the recommended content is displayed by a display method different from the display method when the number of users is two or less. In this embodiment, a circle display is shown as an example (step 603).
FIG. 7 shows a circle display example of step 603, which is displayed on the content recommendation device 105. Reference numerals 701, 702, and 703 are areas in which marks such as icons and images for specifically displaying the users A, B, and C are displayed. These are arranged on the circle 700 at equal intervals. The outside of the circle 700 is radially divided into six areas by dotted lines. Note that the dotted line is not necessarily displayed on the content recommendation device 105. Of these six areas, an area 704, an area 705, and an area 706 including user marks 701, 702, and 703 are areas in which recommended content for the user indicated by the user mark is displayed. That is, an area 704 including a user mark 701 as a user A mark is an area for displaying recommended content for the user A. In this embodiment, one or a plurality of items representing content features such as content thumbnails and titles are displayed in all areas displaying recommended content.

  An area 707 located between the area 704 and the area 705 is an area in which recommended content for the two users A and B is displayed. Similarly, an area 708 located between the area 704 and the area 706 is an area in which recommended content for the two users A and C is displayed. Area 709 is an area in which recommended content for two users, user B and user C, is displayed. In this case, the weighting between the two users is equal, and neither preference is made stronger. This process is calculated by the recommended content calculation unit 304 for user combination.

  An area 710 is a display area that displays recommended content for all users, and displays recommended content that satisfies the preferences of all the users A, B, and C at the same time.

  When the number of users increases beyond a certain level, it is also possible to display in a table format as shown in FIG. 8 without narrowing down the users. Specifically, an area where the row of user C and the column of user A intersect is an area where recommended content for the combination of user A and user C is displayed. In the upper left area, recommended programs for all the users A, B, C, and D are displayed. Thereby, even when the number of persons is four or more, it is possible to present recommended content equivalent to the display content of FIG.

  According to the embodiment described above, it is possible to realize both the presentation of content that matches the preferences of individual users and the presentation of content that matches the common preferences.

  By the above processing, even when the number of users increases, content for all users and individual recommended content can be displayed simultaneously.

  This embodiment is a content recommendation apparatus and method that provide a recommended content calculation method for user combinations different from that of the second embodiment. The configuration block diagram of this embodiment is the same as that of the second embodiment, and FIG. 2 will be used and will be described with reference to FIGS.

In FIG. 3, in the recommended content calculation unit 304 for user combination, two users are treated equally, whereas in FIG. 9 of this embodiment, the recommendation for weighted user combination is used instead of the recommended content calculation unit 304 for user combination. The content calculation unit 901 is used to calculate recommended content that simultaneously satisfies the preferences of two users. (Step 1001)
A weighted user combination recommended content calculation unit 901 weights user preferences when calculating recommended content for two users. Specifically, when user A and user B are weighted at a ratio of 2: 1, even if the preference parameter for the same performer registered in the user information registration unit 201 is the same, user A is better The recommended content is calculated as having twice as much preference as the user B. At the same time, the recommended content when the user A and the user B are weighted at a ratio of 1: 2 is also calculated. Similarly, the recommended content in the case where the weight is large or small is calculated for all combinations of two users from all users specified in step 402. As the weighting ratio, a fixed value set in advance can be used, or a variable value that can be set later by the user can be used. The weighting ratio is stored in the user information registration unit 201 in FIG. 2, and is retained even when the power is turned off. In this way, by weighting the user's preference and calculating the recommended content for the two users, it is possible to calculate the recommended content with a higher degree of preference.

  FIG. 11 shows a display example of step 603 for displaying the recommended content calculated in step 1001 in a circle on the content recommendation device 105.

  The outside of the circle 700 is radially divided into nine areas by dotted lines. Note that the dotted line is not necessarily displayed on the content recommendation device 105. Of these nine areas, 1101 is a recommended content display area that is selected when the degree of preference of the user A is made stronger than that of the user B among the contents satisfying the preferences of the user A and the user B at the same time. Since the degree of preference of the user A is stronger than that of the user B, the area 1101 is arranged on the side closer to the recommended content area for the user A. On the other hand, the recommended content display area selected when the degree of preference of the user B is stronger than that of the user A is arranged in an area 1102 on the side close to the recommended content area for the user B. Similarly, the recommended content for the user A and the user C having a higher degree of preference for the user A is displayed in the area 1103, and the recommended content for the user A and the user C having a higher degree of preference for the user C are displayed in the area 1103. In area 1105, recommended content for user B and user C with a higher degree of preference for user B is shown in area 1105, and recommended content for user B and user C with a higher degree of preference for user C are in area 1105 1106 is displayed.

  It is also possible to display in a table format as shown in FIG. The user component degree of the row component is weighted to be stronger than the user component degree of the column component. Specifically, an area where the row of user A and the column of user B intersect is selected as a recommendation content satisfying both user A's and user B's preferences at the same time because user A's preference is stronger than user B's. The recommended content is displayed. On the contrary, in the area where the row of the user B and the column of the user A intersect, the recommended content selected more strongly reflecting the preference of the user B is displayed.

  According to the embodiment described above, it is possible to realize both the presentation of content that matches the preferences of individual users and the presentation of content that matches the common preferences.

  By the above process, the preference for the number of users can be more reflected, and content for all users and individual recommended content can be efficiently displayed simultaneously.

  This embodiment is a content recommendation apparatus and method that provides processing when the number of users increases to a certain level or more in the second embodiment. The configuration block diagram in the present embodiment is the same as that in the second embodiment, and FIG. 2 will be used and will be described with reference to FIGS.

  When the number of people increases to four or more, there are cases where the display method of FIG. Specifically, the user A, the user B, the user C, and the user D are arranged on the circle at equal intervals in the clockwise direction, and the area is divided on the radiation, and the recommended content for the combination of adjacent users is displayed. Although it is possible, it is not possible to display recommended content for combinations of users A and C and users B and D that are not adjacent to each other. In order to cope with this, it has been shown that when the number of users increases to a certain level or more, it can be solved by the tabular display method as shown in FIG. However, the size of the information display unit 106 is limited, and as the number of users increases, the area that can be displayed for one user is reduced.

In FIG. 13, in order to cope with this, a priority user calculation unit 1301 is newly provided. The priority user calculation unit 1301 selects three users in descending order of priority from the users acquired by the user information acquisition unit 104. (Step 1401)
As the user priority, a fixed value registered in advance by the user in the user information registration unit 201 or a variable value that changes according to the time zone and the situation is set. Even when the user priorities are not registered in advance, it is possible to automatically select three users in the order closer to the center of the image acquired by the user information acquisition unit 104. Further, even after three users are selected once, it is possible to select three again from all users by remote control operation.

  The selected recommended content for three people is displayed as shown in FIGS.

  Even when the number of users increases to four or more by the above processing, the recommended content can be selected without impairing the operability of the user by displaying in the same display format as the display up to the time of three. it can.

  The present embodiment is a content recommendation apparatus and method that provides processing after user identification different from the first embodiment. The configuration block diagram and the functional block diagram in this embodiment will be described with reference to FIGS. 15 and 16 using FIGS. 2 and 3 which are the same as those in the first embodiment.

  In step 1501 of FIG. 15, when it is recognized that there is a user who is not registered in the user information registration unit 201, the subsequent processing is changed.

  The user not registered in the user information registration unit 201 is determined as a guest user, and the recommended content determined from the gender and the age group or the recommended content of the genre for guest set in advance is calculated from the data in the content information storage unit 203 . (Step 1502) For the recommended content for guests, ranking information obtained from the outside via a network or the like may be used.

Only the recommended content calculated in step 1502 is displayed on the content recommendation device 105. (Step 1503)
The reason why only the recommended program guide for the guest user is displayed in step 1503 is that the case where the guest does not want to know the user's preference from the recommended content is considered.

  FIG. 16 shows an example displayed on the content recommendation device 105 in step 1503. When it is determined that there is a guest user 1601, only the recommended content for the guest user is displayed in the area 1602 on the content recommendation device 105.

  With the above processing, it is possible to display the recommended content corresponding to the case where there is a user whose user information is not registered, and it is possible to protect the privacy of the registered user.

101, 102 ... User A, User B
103 ... Remote control
104 ... User information acquisition unit
105 ・ ・ ・ Content recommendation device
106 ... Information display section on content recommendation device
501, 502, 503... Recommended content display area for user A, user B, user A and user B
701, 702, 703 ... Marks for identifying user A, user B, and user C
704, 705, 706... Recommended content display area for user A, user B, user C
707, 708, 709 ... Recommended content display area for user A user B, user A user C, user B user C
710 ... Recommended content display area for user A user B user C
1101... Recommended content display area for user A and user B that enhances user A's preference
1102... Recommended content display area for user A and user B that enhances user B's preference
1103... Recommended content display area for user A and user C with stronger user A preference
1104... Recommended content display area for user A and user C with enhanced user C preference
1105... Recommended content display area for user B and user C that has enhanced user B's preference
1106... Recommended content display area for user B and user C with enhanced user C preference
1601 ・ ・ ・ Guest user
1602 ・ ・ ・ Recommended content display area for guest users

Claims (5)

A content recommendation device that performs display for recommending content,
A user information acquisition unit for recognizing a user and acquiring user information;
A user information registration unit that pre-stores content preference information for each user;
A content information storage unit for storing information related to content preference information;
A display unit for displaying content recommendations;
When there are a plurality of users recognized by the user information acquisition unit, the content that is preferred for each user and each combination of a plurality of users is selected and selected using the user information registration unit and the content information storage unit. A content recommendation device comprising: a control unit that simultaneously displays the content on a display unit for each user and each combination of a plurality of users. The content recommendation device according to claim 1,
The content recommendation apparatus, wherein the control unit changes a display format on the display unit according to the number of users recognized by the user information acquisition unit. The content recommendation device according to claim 1,
The content recommendation device, wherein the control unit changes display contents on the display unit according to weighting of a plurality of users recognized by the user information acquisition unit. The content recommendation device according to claim 1,
When the number of users recognized by the user information acquisition unit is greater than or equal to a threshold value, the control unit narrows the user to the threshold value or less to select content and display the content. The content recommendation device according to claim 1,
If the information related to the user recognized by the user information acquisition unit is not held in the user information registration unit, the control unit is a content selected based on the user information acquired by the user information acquisition unit. Alternatively, a content recommendation device that displays preset content on the display unit.

Images