JP2014041557A - Content recommendation system, terminal device, server, content recommendation method, and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a content recommendation system for providing a content in accordance with a user's preference, wherein a use state of a terminal device in a mutually disconnected state with a server can be reflected to estimation of the user's preference.SOLUTION: In a content recommendation system including a terminal device and a server, communicating with each other via a network, the terminal device generates terminal state information including a content of an operation to an operation input part so as to transmit it to the server. The server registers received terminal state information in terminal state history stored in a storage part. Also, the server determines, on the basis of the terminal state history, a recommendation content to be recommended to a user at the terminal device out of recommendation candidate contents, and transmits a recommendation content list indicative of a determined recommendation content. The terminal device causes a display part to display the recommendation content list received from the server.

Description

  The present invention relates to a content recommendation system, a terminal device, a server, a content recommendation method, and a program.

There is known a system that recommends content that matches a user's preference to a terminal device capable of communication via a network.
For example, Patent Document 1 discloses a system in which communication between a client and a server is monitored and an advertisement is distributed according to user preferences (interaction frequency and occurrence time). Specifically, in this system, for example, the server monitors interaction messages by monitoring phone messages, downloaded applets, selected RSS feeds and search terms or navigation links used in the Wireless Application Protocol (WAP), etc. Based on the frequency and occurrence time, an advertisement according to the user's preference is distributed.

Special table 2011-525257 gazette

As a specific example, if the user is looking at the content list and is wondering which content to download at this time, the list of content that the user is browsing at that time will not be downloaded after all It also suits user preferences. Even if the content is such that the user is confused about downloading, it is preferable that the content of the recommended content is enhanced if it is reflected on the user's preference on the server side.
However, when browsing the content list in this way, the operation according to the operation for that is completed by the terminal device, and does not involve mutual communication with the server. In the system of Patent Document 1, the user's preference is estimated from the result of a terminal device that is a client performing communication with a server and executing some processing. Therefore, depending on Patent Document 1, even if the terminal device that reflects the user's preference is used, the user device's preference is not obtained for the case where the terminal device and the server are not interconnected. It cannot be reflected in the estimation.

  The present invention has been made in view of such circumstances, and in a content recommendation system that provides content according to the user's preference, the usage status of the terminal device in a state in which the terminal device is not interconnected with the user's preference. It is intended to be able to be reflected in the estimation of.

  In order to solve the above-described problems, a content recommendation system as one aspect of the present invention includes a terminal device and a server that communicate with each other via a network, and the terminal device includes an operation input unit that inputs an operation; A terminal state information processing unit that generates terminal state information including the content of the operation input by the operation input unit as information indicating a predetermined state of the terminal device, and transmits the generated terminal state information to the server; A recommended content list control unit that executes at least control for causing the display unit to display the recommended content list received from the server, and the server registers the terminal state information transmitted from the terminal device in the terminal state history A terminal status history management unit that performs the terminal status history, a server storage unit that stores the terminal status history, and a recommendation candidate configuration based on the terminal status history. Content recommendation system and a recommended content determining unit that determines a recommended content to be recommended to the user of the terminal device from among Ntsu transmits the recommended content list indicating the determined recommended content to the terminal device.

  In the content recommendation system of the present invention, the recommended content determination unit determines an associative word search word that is a character string for searching for an associative word based on the terminal state history. And an associative word search unit that searches for an associative word based on the associative word search word, and a recommended content that is a character string for searching for recommended content based on the associative word searched by the associative word search unit A recommended content search term determination unit that determines a search term for use, and a recommended content search unit that determines, as the recommended content, content searched from recommended candidate content based on the search term for recommended content. Good.

  In the content recommendation system of the present invention, the associative word search term determination unit includes a content selection unit that selects content that matches the user's preference based on the terminal state history, and the content selected by the content selection unit. A morpheme analysis unit that performs a morpheme analysis on a character string included in the meta information, and a morpheme correspondence that determines a morpheme selected from among the morphemes analyzed by the morpheme analysis unit as at least one of the association word search words And a determination unit.

  In the content recommendation system of the present invention, the terminal state history stores operation position history information indicating a history of positions operated on the screen, and the associative word search word determination unit includes the operation position history. You may provide the operation position corresponding | compatible determination part which determines at least 1 of the said search words for associative words based on information.

  In the content recommendation system of the present invention, the terminal status history stores operation time zone history information indicating a history of a time zone in which a user performs an operation on the terminal device, and determines the search term for the associative word The unit may include an operation time correspondence determination unit that determines at least one of the associative word search words based on the operation time zone history information.

  The terminal device according to one aspect of the present invention generates an operation input unit that inputs an operation, and terminal state information including the content of the operation input by the operation input unit as information indicating a predetermined state of the terminal device. A terminal state information processing unit that transmits the generated terminal state information to the server, and a recommendation that executes at least control for displaying the recommended content list received from the server that communicates with the terminal device via the network on the display unit And a content list control unit.

  In addition, a server as one aspect of the present invention stores a terminal state history management unit that registers terminal state information transmitted from a terminal device that communicates with a server via a network in a terminal state history, and the terminal state history Based on the server storage unit and the terminal state history, the recommended content to be recommended to the user of the terminal device is determined from the recommended candidate content, and the recommended content list indicating the determined recommended content is stored in the terminal device. And a recommended content determination unit to be transmitted.

  A content recommendation method as one aspect of the present invention is a content recommendation method in a content recommendation system including a terminal device and a server that communicate with each other via a network, and the terminal device is an operation for inputting an operation. Terminal state information processing for generating terminal state information including the content of the operation input by the operation input step as information indicating an input step and a predetermined state of the terminal device, and transmitting the generated terminal state information to the server And a recommended content list control step for executing at least control for displaying a recommended content list received from the server on a display unit, wherein the server receives the terminal status information transmitted from the terminal device as a server A terminal state history management step for registering in the terminal state history stored in the storage unit; Based on the terminal state history, a recommended content to be recommended to a user of the terminal device is determined from recommended candidate content, and the recommended content list indicating the determined recommended content is transmitted to the terminal device. Steps.

  The content recommendation method as one aspect of the present invention includes an operation input step for inputting an operation, and terminal state information including the details of the operation input by the operation input step as information indicating a predetermined state of the terminal device. A terminal state information processing step of generating and transmitting the generated terminal state information to the server, and at least control for displaying the recommended content list received from the server communicating with the terminal device via the network on the display unit A recommended content list control step.

  The content recommendation method as one aspect of the present invention includes a terminal state history management step of registering terminal state information transmitted from a terminal device communicating with a server via a network in a terminal state history stored in a server storage unit Based on the terminal status history, the recommended content to be recommended to the user of the terminal device is determined from the recommended candidate content, and the recommended content list indicating the determined recommended content is transmitted to the terminal device. A content determination step.

  Further, the program as one aspect of the present invention is a terminal state information including an operation input step for inputting an operation to a computer and an operation content input by the operation input step as information indicating a predetermined state of the terminal device. At least a terminal state information processing step for transmitting the generated terminal state information to the server, and a control for displaying a recommended content list received from the server communicating with the terminal device via the network on the display unit The recommended content list control step is executed.

  Further, the program as one aspect of the present invention is a terminal state history management that registers terminal state information transmitted from a terminal device that communicates with a server via a network in a terminal state history stored in a server storage unit. Determining recommended content to be recommended to the user of the terminal device from recommended candidate content based on the terminal status history and transmitting the recommended content list indicating the determined recommended content to the terminal device And a recommended content determination step.

  As described above, according to the present invention, in the content recommendation system that provides content according to the user's preference, the usage status of the terminal device in a state of not being interconnected with the server is reflected in the estimation of the user's preference. The effect that it becomes possible to obtain is obtained.

It is a figure which shows the structural example of the content recommendation system in this embodiment. It is a figure which shows the operation example for the content recommendation in the content recommendation system of this embodiment. It is a figure which shows the function structural example of the terminal device and server system in this embodiment. It is a figure which shows the example of the content of the preference analysis information database in this embodiment. It is a figure which shows the example of the aspect of the screen of the content browsing application in this embodiment. It is a figure which shows the management example of the operation position information according to operation with respect to a content list area | region. It is a figure which shows the example of management of the operation position information according to operation with respect to a service list area | region. It is a figure which shows the management example of the operation position information according to operation with respect to a download list area | region. It is a figure which shows the structural example of the terminal state information in this embodiment. It is a figure which shows the structural example of the content information in this embodiment. It is a figure which shows the structural example of the service information in this embodiment. It is a figure which shows the structural example of the user information table in this embodiment. It is a figure which shows the structural example of the terminal status log | history table in this embodiment. It is a figure which shows the example of a division | segmentation setting of the screen of the content browsing application in this embodiment. It is a figure which shows the example of a division | segmentation setting of the screen of the content browsing application in this embodiment. It is a figure which shows the example of a division | segmentation setting of the screen of the content browsing application in this embodiment. It is a figure which shows the structural example of the search word table in this embodiment. It is a figure which shows the structural example of the content browsing condition log | history table in this embodiment. It is a figure which shows the example of the content of the search word table according to the example of determination of the search word for associative words in this embodiment. It is a figure which shows the example of the content of the search word table according to the example of determination of the search word for recommended content in this embodiment. Associative words determined based on the result of searching the associated word only by the associated word search word determined based on the content that matches the user's preference, and the operation result performed by the user on the screen of the content browsing application It is a figure which compares and shows the result which searched the associative word using the search word together. It is a figure which shows the example of the content of the content meta information storage part in this embodiment. It is a figure which shows the structural example of the book entry table in this embodiment. It is a figure which shows the structural example of the book information table in this embodiment. It is a figure which shows the structural example of the recommended content list | wrist in this embodiment. It is a figure which shows the example of a process sequence which a terminal device, a preference analysis server, and a content delivery server in this embodiment perform for content recommendation. The figure which shows the example of a procedure of the recommendation content list production | generation process which the recommendation content determination part of the preference analysis server in this embodiment performs, and the process which a content delivery server, a content management server, and an associative word search server perform in connection with this process is there.

[Configuration example of content recommendation system]
FIG. 1 shows a configuration example of a content recommendation system in an embodiment of the present invention. Note that the type of content to be recommended in the content recommendation system of the present embodiment is not particularly limited, but in the following description, a case where an electronic book is the content to be recommended is taken as an example.

  The content recommendation system of this embodiment includes a terminal device 100 and a server system (server) 200 for each user. The terminal device 100 and the server system 200 communicate via the network 300.

The terminal device 100 is a network terminal owned by a user. In order to acquire a list of recommended books (recommended content list), the terminal device 100 in the present embodiment transmits terminal state information indicating a predetermined state about itself to the server system 200. This terminal state information includes a result of an operation performed on the terminal device 100 by the user, for example.
Also, the terminal device 100 can acquire a recommended content list from the server system 200 and display the acquired recommended content list. Specifically, the recommended content list is a list of electronic books selected according to the user's preference of the terminal device 100.
Note that specific examples of the terminal device 100 include a personal computer, a mobile phone, a smartphone, and the like.

The server system 200 registers the received terminal state information in the terminal state history each time terminal state information is received from the terminal device 100. The server system 200 stores the terminal state history in association with each user. This terminal state history includes a history (log) of operations performed on the terminal device 100 by the user.
The server system 200 determines an electronic book (content) to be recommended to the user based on the terminal state history. That is, the server system 200 determines an electronic book that matches the user's preference. Then, the server system 200 transmits the electronic book list thus determined to the terminal device 100 as a recommended content list.

  The server system 200 of this embodiment may be configured by including a plurality of servers. In the example of FIG. 1, the server system 200 includes a content distribution server 201, a content management server 202, a terminal state information management server 203, a preference analysis server 204, and an associative word search server 205.

  The content distribution server 201 is a server that distributes content as an electronic book to the terminal device 100.

The content management server 202 is a server that manages electronic books owned by the user of the terminal device 100. Specifically, the content management server 202 manages meta information of an electronic book owned by a user as book shelf information for each user.
For example, information on a bookshelf of a certain user (information on an electronic book owned by the user) can be browsed by another user accessing the content management server 202 by the terminal device 100. In addition, the content management server 202 determines recommended content in response to a request from the preference analysis server 204, and transmits a recommended content list for the determined recommended content to the preference analysis server 204.

  The terminal state information management server 203 transfers the terminal state information transmitted from the terminal device 100 to the preference analysis server 204. At this time, the terminal device 100 can convert the format of the terminal state information transmitted from the terminal device 20 as necessary, for example, and transmit it to the preference analysis server 204.

The preference analysis server 204 registers the terminal state information received from the terminal state information management server 203 in the terminal state history. In addition, the preference analysis server 204 executes a process for acquiring a recommended content list based on the terminal state history.
The associative word search server 205 searches for an associative word in response to a request from the preference analysis server 204.

[Operation example of content recommendation system]
An example of the operation of the content recommendation system of this embodiment will be described with reference to FIG.
The terminal device 100 generates terminal state information indicating its own state according to the use of the user, and transmits the generated terminal state information to the terminal state information management server 203 at a predetermined timing (step S1).

(Terminal status information transmission timing)
For example, the terminal device 100 may determine a transmission time in advance as the transmission timing of the terminal state information, and transmit the terminal state information at a timing when the transmission time is reached. At this time, the terminal device 100 retains the generated terminal state information until the transmission time is reached, and transmits the terminal state information that has been retained so far at the timing when the transmission time is reached. Further, when the user is operating, terminal state information generated per unit time increases, so the terminal device 100 may transmit the terminal state information at regular intervals.
Note that the terminal device 100 may transmit the generated terminal state information each time the terminal state information is generated according to, for example, a user operation. However, in this case, communication processing frequently occurs when the user is operating the terminal device 100, and the processing load for communication becomes heavy. For this reason, it is preferable to transmit at predetermined transmission times as described above, or to transmit at regular intervals.

  The terminal state information management server 203 transfers the terminal state information received from the terminal device 100 to the preference analysis server 204 (step S2). The terminal state information management server 203 may store terminal state information received from the terminal device 100.

  The preference analysis server 204 registers the terminal state information transmitted in step S2 in the terminal state history. The preference analysis server 204 executes control for acquiring a recommended content list each time a certain amount of terminal state history is accumulated. In obtaining the recommended content list, the preference analysis server 204 requests the content distribution server 201 for meta information (content information) of the content specified by the terminal state history (step S3).

  The content distribution server 201 stores content for distribution to the terminal device 100. The content distribution server 201 selects content specified by the terminal state history from the distribution content. Then, the meta information (content information) of the selected content is transmitted to the preference analysis server 204 (step S4).

  The preference analysis server 204 receives the content information transmitted in step S4. This content information includes, for example, character string information including the title, author, and commentary of the electronic book that is the content. In the present embodiment, the content may be content corresponding to a region. The content information in this case also includes character string information related to the corresponding region.

The preference analysis server 204 performs a morphological analysis on the character string extracted from the content information, and determines an associative word search word using a result of the morphological analysis.
Further, the terminal state information includes information on a position where an operation is performed on the touch panel of the terminal device 100. The preference analysis server 204 also determines a search term for an associative word based on a history of positions at which operations are performed on the touch panel of the terminal device 100 indicated in the terminal state information history.
Further, the terminal state information includes information on the time when the operation is performed on the terminal device 10. The preference analysis server 204 also determines the search term for the associated word based on the history of the time when the operation was performed on the terminal device 100 indicated in the terminal state information history.
Then, the preference analysis server 204 transmits an associative word search request to the associative word search server 205 together with the associative word search word determined as described above (step S5).

  The associative word search server 205 searches for an associative word corresponding to the associative word search word in response to the associative word search request transmitted in step S5, and transmits the search result to the preference analysis server 204 (step S6). ).

  Next, the preference analysis server 204 uses the associative word search result received in step S6 to determine a recommended content search word to be used for searching recommended content. The preference analysis server 204 transmits a recommended content list request to the content management server 202 together with the determined recommended content search word (step S7).

In response to the recommended content list request, the content management server 202 searches for electronic books corresponding to the recommended content search word received together with the recommended content list request from among the electronic books stored by itself. Then, information indicating the retrieved electronic book is transmitted to the preference analysis server 204 as a recommended content list (step S8).
The preference analysis server 204 transfers the recommended content list received in step S8 to the content distribution server 201 (step S9).

The content distribution server 201 stores the recommended content list transferred from the preference analysis server 204. Then, in response to a request for a recommended content list from the terminal device 100, the stored recommended content list is transmitted to the requesting terminal device 100 (step S10).
Here, the recommended content list transferred from the preference analysis server 204 is, for example, in XML (eXtensible Markup Language) format. Therefore, the content distribution server 201 converts the recommended content list in the XML format into an RSS (RDF site summary) feed format or an atom feed format so that the content can be viewed on the terminal device 100, and then the terminal device 100. Send to.
Note that the transmission of the recommended content list from the content distribution server 201 to the terminal device 100 may be performed by the content distribution server 201 by a push type.

  The terminal device 100 stores the recommended content list received in step S10. Then, the terminal device 100 displays the stored recommended content list on the display unit in response to an operation for instructing the user to display the recommended content list. Thereby, the user can browse the list of electronic books recommended according to his / her preference.

[Functional configuration example of terminal device]
FIG. 3 shows a functional configuration example of the terminal device 100 and the server system 200. In this figure, the terminal state information management server 203 is not shown.
First, a functional configuration example of the terminal device 100 will be described with reference to FIG.
The terminal device 100 includes an operation input unit 101, a terminal state information processing unit 102, a recommended content list control unit 103, a terminal storage unit 104, and a display unit 105.

  The operation input unit 101 inputs an operation to be performed on an operator or an operation device provided in the terminal device 100. Note that the operation device included in the terminal device 100 includes a touch panel integrated with the display unit 105.

  The terminal state information processing unit 102 generates terminal state information including the content of the operation input by the operation input unit 101 as information indicating a predetermined state of the terminal device 100. The terminal state information processing unit 102 transmits the generated terminal state information to the server system 200.

  The recommended content list control unit 103 executes at least control for causing the display unit 105 to display the recommended content list received from the server system 200. Also, the recommended content list control unit 103 stores the recommended content list received from the server system 200 in the terminal storage unit 104. When displaying the recommended content list on the display unit 105, the recommended content list control unit 103 reads the recommended content list stored in the terminal storage unit 104 and displays the recommended content list on the display unit 105.

  The terminal storage unit 104 stores various information used in the terminal device 100 including a recommended content list.

  The display unit 105 displays a recommended content list according to the control of the recommended content list control unit 103. Also, other user interface images can be displayed as appropriate according to user operations.

[Functional configuration example of terminal device]
Next, a configuration example of the server system 200 will be described with reference to the same FIG.
The server system 200 includes a terminal state history management unit 210, a server storage unit 220, and a recommended content determination unit 230.

  The terminal state history management unit 210 registers the terminal state information transmitted from the terminal device 100 in the terminal state history. In the present embodiment, the terminal state history management unit 210 is provided in the preference analysis server 204.

  The server storage unit 220 is provided in the preference analysis server 204. The server storage unit 220 stores the terminal state history in which the terminal state information is registered as described above. Note that the server storage unit 220 of this embodiment stores a preference analysis information database 221 as shown in FIG. The terminal state history is stored in the preference analysis information database 221.

The recommended content determination unit 230 determines recommended content to be recommended to the user of the terminal device from the recommended candidate content based on the terminal state history, and provides a recommended content list indicating the determined recommended content to the terminal device 100. To send.
In the present embodiment, the function as the recommended content determination unit 230 is distributed to the preference analysis server 204, the content management server 202, and the associative word search server 205.
In addition, when transmitting the recommended content list to the terminal device 100, the recommended content list may be transmitted directly to the terminal device 100 or indirectly. In the case of the present embodiment, the recommended content determination unit 230 indirectly transmits the recommended content list to the terminal device 100 via the recommended content list distribution unit 281 provided in the content distribution server 201.

The recommended content determination unit 230 includes an associative word search word determination unit 240, an associative word search unit 250, a recommended content search word determination unit 260, and a recommended content search unit 270.
The associative word search word determination unit 240 determines an associative word search word that is a character string for searching for an associative word based on the terminal state history. In the present embodiment, the associative word search word determination unit 240 is provided in the preference analysis server 204.

The associative word search unit 250 searches for an associative word based on the associative word search word. At this time, the associative word search unit 250 searches the associative word database 251 included in the associative word search server 205 for an associative word corresponding to the associative word search word.
In the present embodiment, the associative word search unit 250 is provided in the associative word search server 205.

  The recommended content search word determination unit 260 determines a recommended content search word that is a character string for searching for recommended content based on the associative word searched by the associative word search unit 250. In this embodiment, the function of the recommended content search word determination unit 260 is provided in the preference analysis server 204.

The recommended content search unit 270 determines, as recommended content, the content searched from the recommended candidate content based on the recommended content search terms. At this time, the recommended content search unit 270 searches the content meta information storage unit 271 included in the content management server 202 for content having meta information corresponding to the recommended content search word.
In the present embodiment, the recommended content search unit 270 is provided in the content management server 202.

  The associative word search term determination unit 240 further includes a content selection unit 241, a morpheme analysis unit 242, a morpheme correspondence determination unit 243, an operation position correspondence determination unit 244, and an operation time correspondence determination unit 245.

The content selection unit 241 selects content that matches the user's preference based on the terminal state history. The content selection unit 241 also selects a service that matches the user's preference based on the terminal state history.
In the configuration of FIG. 3, the content selection unit 241 requests content information or service information from the content distribution server 201 together with the ID (content ID or service ID) of the selected content or service.
In response to this request, the content search unit 283 of the content distribution server 201 reads from the distribution content storage unit 282 content information or service information that is meta information about the content indicated by the content or service ID received together with the request. The content search unit 283 transmits the read content information or service information to the preference analysis server 204. Accordingly, the content selection unit acquires meta information about the selected content or service.

The morphological analysis unit 242 performs a morphological analysis on the character string included in the meta information of the content selected by the content selection unit 241.
The morpheme correspondence determination unit 243 determines a morpheme selected from the morphemes analyzed by the morpheme analysis unit 242 as at least one of the association word search words.

  The operation position correspondence determining unit 244 determines at least one of the search words for associative words based on operation position history information indicating a history of positions operated on the screen.

  The operation time correspondence determining unit 245 determines at least one of the associative word search terms based on operation time zone history information indicating a history of a time zone in which the user performed an operation on the terminal device.

FIG. 4 shows an example of the contents of the preference analysis information database 221 stored in the server storage unit 220.
The preference analysis information database 221 includes a user information table 510, a terminal state history table 520, a search term table 530, and a content browsing status history table 540.

The user information table 510 stores predetermined information related to the user for each user.
The terminal state history table (terminal state history) 520 is a table that stores various types of information as terminal state history in association with each user.
The search word table 530 stores, for each user, the associative word search words determined by the associative word search word determination unit 240 and the recommended content search words determined by the recommended content search word determination unit 260. .
The content browsing status history table 540 is a table created based on the terminal status information, and stores a history of browsing status of user content for each user.

[Content browsing application screen example]
In the present embodiment, the terminal state information transmitted from the terminal device 100 to the preference analysis server 204 in steps S1 and S2 of FIG. 2 is the operation performed on the screen of the content browsing application displayed on the display unit 105. Reflected.

FIG. 5 shows a screen example of the content browsing application of the present embodiment displayed on the display unit 105.
The screen of the content browsing application shown in FIG. 5A is divided into a service list area AR1, a content list area AR2, and a download list area AR3 from top to bottom.

  The service list area AR1 is an area for displaying a list of services that can be browsed. Each thumbnail image TN1 arranged in the horizontal direction in the service list area AR1 indicates one service. When selecting a service, the user performs a tap operation on the thumbnail image TN1 corresponding to the service to be selected from the thumbnail images TN1 in the service list area AR1.

The content list area AR2 is an area in which a list of contents (electronic books) provided in the service selected by the tap operation on the service list area AR1 is displayed.
In the content list area AR2, for example, thumbnail images TN2 of the cover of the electronic book provided in the selected service are displayed in an array. The breakdown of the cover of the electronic book displayed in the content list area AR2 differs for each selected service.
For example, when the user performs a tap operation on the thumbnail image TN2 displayed in the content list area AR2, the display is switched to the electronic book view display of the thumbnail image TN2 for which the tap operation has been performed.

The download list area AR3 is an area in which a list of contents (electronic books) downloaded from the download list by the user is displayed. In the content list area AR2, for example, thumbnail images TN3 of the cover of a downloaded electronic book are arranged and displayed.
For example, when the user performs a tap operation on the thumbnail image TN3 displayed in the download list area AR3, the display is switched to a state where the electronic book of the thumbnail image TN3 on which the tap operation has been performed can be viewed.

  The download list area AR3 can be hidden on the screen as shown in FIG. 5B, for example, by a predetermined operation on the screen.

[Example of management of operation information for content browsing application screens]
Each time a tap operation is performed on the screen illustrated in FIG. 5, the terminal state information processing unit 102 of the terminal device 100 according to the present embodiment generates operation state information with the tap operation as one event.
Therefore, an example of management of operation information for the screen of the content browsing application by the terminal state information processing unit 102 will be described with reference to FIGS.

First, a management example of operation position information corresponding to an operation on the content list area AR2 will be described with reference to FIG.
The thumbnail display position Dpos2 for displaying the thumbnail image TN2 in the content list area AR2 is defined as 6 rows × 3 columns as shown in FIG. 6A, for example. The terminal state information processing unit 102 sets display position numbers 1 to 18 for each of these 18 thumbnail display positions Dpos2.
Further, the terminal state information processing unit 102 arranges a list of contents cnt in one service in correspondence with the arrangement of thumbnail display positions Dpos2 as shown in FIG. 6B, for example. In addition, the terminal state information processing unit 102, in accordance with the arrangement order of these contents cnt, shows the position (list position) number (list position) in the entire contents list for each content cnt, as indicated by the numbers shown in circles in the figure. Position number).

  In the content list area AR2, the thumbnail images TN2 are displayed in the arrangement order of the content cnt shown in FIG. When the number of contents cnt is larger than the number of thumbnail display positions Dpos2, the user performs a scrolling operation in the vertical direction to display thumbnail images TN2 of the contents cnt arranged as shown in FIG. Can be moved to.

In FIG. 6B, the thumbnail image TN2 of the contents cnt whose list position numbers are Nos. 4 to 21 with respect to the thumbnail display positions Dpos2 whose display position numbers are Nos. 1 to 18 in FIG. 6A, respectively. Is displayed.
At this time, it is assumed that the user performs a tap operation on the thumbnail image TN2 displayed at the thumbnail display position Dpos2 of the seventh display position number. The list position number of the content cnt to which the thumbnail image TN2 displayed at the 7th display position number corresponds is the 10th.
In this case, the terminal state information processing unit 102 generates a combination of the 7th display position number and the 10th list position number as operation position information in response to an event as a tap operation on the content list area AR2.

Next, an example of management of operation information corresponding to an operation on the service list area AR1 will be described with reference to FIG.
For the display position number of the thumbnail display position Dpos1 where the thumbnail image TN1 is displayed in the service list area AR1, the terminal state information processing unit 102 moves from left to right in the service list area AR1, as shown in FIG. 7A, for example. Set numbers 1 to 5 in order.
Further, the terminal state information processing unit 102 arranges a list of services srv in the horizontal direction as shown in FIG. 7B, for example. In addition, the terminal state information processing unit 102 sets the list position number in the entire service list for each service srv, as indicated by the numbers in circles in the figure, according to the arrangement order of these services srv.

  In the service list area AR1, the thumbnail images TN1 for each service srv are displayed in the order of arrangement of the services srv in FIG. The user can move the thumbnail image TN1 of the service srv arranged in the horizontal direction, for example, as shown in FIG.

In FIG. 7B, the thumbnail images TN1 of the services whose list position numbers are 2 to 6 are respectively displayed with respect to the thumbnail display positions Dpos1 whose display position numbers are 1 to 5 in FIG. 7A. Indicates the displayed state.
At this time, it is assumed that the user performs a tap operation on the thumbnail image TN1 displayed at the thumbnail display position Dpos1 of the third display position number. The list position number of the service srv to which the thumbnail image TN1 displayed at the third display position number corresponds is the fourth.
In this case, the terminal state information processing unit 102 generates a combination of the third display position number and the fourth list position number as operation position information in response to an event as a tap operation on the service list area AR1.

Next, a management example of operation information corresponding to an operation on the download list area AR3 will be described with reference to FIG.
For the display position number of the thumbnail display position Dpos3 at which the thumbnail image TN3 is displayed in the download list area AR3, the terminal state information processing unit 102 moves from left to right of the download list area AR3, for example, as shown in FIG. Set numbers 1 to 5 in order.
Further, the terminal state information processing unit 102 arranges the list of contents cnt in the horizontal direction as shown in FIG. 8B, for example. In addition, the terminal state information processing unit 102 sets the list position number in the entire service list for each content cnt, as indicated by the numbers in circles in the figure, in accordance with the arrangement order of these contents cnt.

  In the download list area AR3, the thumbnail images TN3 for each content cnt are displayed in the arrangement order of the content cnt in FIG. 8B. The user can move the thumbnail images TN3 of the contents cnt arranged as shown in FIG. 8B, for example, in the horizontal direction by performing a horizontal scroll operation.

In FIG. 8B, the thumbnail images TN3 of the services whose list position numbers are 2 to 6 are respectively displayed with respect to the thumbnail display positions Dpos1 whose display position numbers are 1 to 5 in FIG. 8A. Indicates the displayed state.
At this time, it is assumed that the user performs a tap operation on the thumbnail image TN3 displayed at the thumbnail display position Dpos1 of the fourth display position number. The list position number of the content cnt corresponding to the thumbnail image TN3 displayed at the fourth display position number is fifth.
In this case, the terminal state information processing unit 102 generates a combination of the fourth display position number and the fifth list position number as operation information in response to an event as a tap operation on the download list area AR3.

[Structure example of terminal status information]
FIG. 9 shows an example of the structure of the terminal state information 410 transmitted from the terminal device 100 to the preference analysis server 204 in steps S1 and S2 of FIG.
The terminal status information 410 includes parameters of info_type 411, event_start_date 412, event_end_date 413, user_ID 414, content_id (service_id) 415, list_kind 416, disp_pos 417, list_pos 418, disp_DL_list 419, and view_kind 420.

info_type 411 indicates a type as terminal state information.
As a specific example, when info_type 411 is “0”, the terminal state information is information related to a tap operation performed on the screen of the content browsing application displayed on the display unit 105 by the user. When the info_type 411 is “1”, the terminal state information is information related to the browsing status displayed on the display unit 105.

The event_start_date 412 indicates the date / time when the event was acquired or the date / time when browsing started.
event_end_date 413 indicates the date and time when browsing as an event was terminated.
The user_ID 414 is a user ID that uniquely identifies a user who uses the terminal device 100 that has transmitted the terminal state information 410.

  The content_id (service_id) 415 indicates a content ID for uniquely identifying the specified content when the content is specified by an operation on the screen of the display unit 105, for example. Alternatively, when a service is designated by an operation on the screen of the display unit 105, a service ID for uniquely identifying the designated service is indicated.

The list_kind 416 indicates the type (list type) of a list that is an operation target when performing an operation on the screen of the content browsing application.
For example, the list_kind 416 is set to a value of “0” indicating that the list type is a service list when an operation is performed on the service list area AR1. The list_kind 416 is set to a value “1” indicating that the list type is a content list when an operation is performed on the content list area AR2. The list_kind 416 is set to a value “2” indicating the download list when an operation is performed on the download list area AR3.

disp_pos 417 is a parameter stored correspondingly when info_type 411 is “0”, and indicates the display position number where the tap operation has been performed in the list area of the type indicated by list_kind 416.
The list_pos 418 is a parameter stored correspondingly when the info_type 411 is “0”, and indicates the list position number where the tap operation is performed in the list area of the type indicated by the list_kind 416.

  The disp_DL_list 419 indicates whether the download list area AR3 is in a display state or a non-display state. For example, when disp_DL_list 419 is “0”, it indicates that the download list area AR3 is displayed. When disp_DL_list 419 is “1”, it indicates that the download list area AR3 is not displayed.

The view_kind 420 indicates whether the browsing screen is in a preview display or a view display when browsing an electronic book as content. For example, when view_kind 420 is “0”, it indicates that the display is a preview display. When view_kind 420 is “1”, it indicates that the display is view display.
The preview display displays, for example, an image for knowing the outline of the electronic book to be browsed. For example, the title of the electronic book and a brief description are displayed together with the cover image of the electronic book.
The view display is displayed in such a manner that the electronic book to be browsed can be read. In view display, for example, an image for one page or an image for two pages of spread is displayed using a large area of the screen so that characters are easy to read. In view display, a page can be turned by an operation such as swipe.

  Further, FIG. 9 shows terminal state information 410A and 410B. Each of the terminal status information 410A and 410B is a CSV file of the terminal status information generated in response to a certain event. In the terminal status information 410A and 410B, character strings are arranged in the order of info_type 411, event_start_date 412, event_end_date 413, user_ID 414, content_id (service_id) 415, list_kind 416, disp_pos 417, list_pos 418, disp_DL_list 419, and view_kind 420. ,, ".

  Each time the operation on the screen of the content browsing application is performed, the terminal state information processing unit 102 generates terminal state information 410 corresponding to the operation. The terminal state information processing unit 102 holds the generated terminal state information. Then, the terminal state information processing unit 102 transmits the held terminal state information 410 to the preference analysis server 204 at steps S1 and S2 at the transmission timing described in FIG.

[Content Information]
Next, content information that the preference analysis server 204 acquires from the content distribution server 201 in steps S3 and S4 shown in FIG. 2 will be described.
First, the content selection unit 241 in the preference analysis server 204 transmits a content information request corresponding to step S3. This content information request stores the content_id (service_id) 415 stored in the terminal state information 410 as information specifying the content.

  The content search unit 283 of the content distribution server 201 searches the content corresponding to the content_id (service_id) 415 stored in the content information request from the content stored in the distribution content storage unit 282. The content search unit 283 reads meta information about the searched content from the distribution content storage unit 282. The content search unit 283 transmits the read meta information corresponding to step S4 to the preference analysis server 204 as content information.

  FIG. 10A shows a structure example of content information 420A acquired by the preference analysis server 204 in step S4 of FIG. The structure shown in this figure is generated by the content search unit 283 in response to content_id 415 stored in the content information request.

FIG. 10A shows an example of the structure of the content information 420A in the XML format.
In this case, the content information 420A includes <id_kind>, <author>, <brand>, <category>, <genre>, <magazineinfo>, <product>, <publisher as main tags in the <entry> tag. >, <Series>, and <bookinfo> are defined. For <bookinfo>, subtitles of <title> and <description> are defined.

The main tag of <id_kind> is the type of ID, but in the present embodiment, it is fixed at <DATA>.
The <author> main tag indicates the author name of the content.
The <brand> main tag indicates the brand name of the content.
The main tag of <category> indicates a category name.
The main tag <genre> indicates the genre of the content.
The main tag of <magazineinfo> indicates a magazine name when the electronic book as content is a magazine.
The main tag of <product> indicates the price of the content.
The main tag of <publisher> indicates the publisher name.
The main tag of <series> indicates the series name when the electronic book as content is serialized.
The main tag of <bookinfo> indicates information about the book.
<Title>, which is a sub tag of <bookinfo>, indicates the title of the content. <Description>, which is a sub tag of <bookinfo>, indicates a description of the content.
FIG. 10B shows a specific example of content information 420A in XML format having the structure shown in FIG.

  If service_id is included in the content information request transmitted in step S3, service information 420B having the structure shown in FIG. 11 is generated by the content search unit 283 instead of the content information 420A, and the preference analysis server 204. Note that the service information 420B in FIG. 11 also shows an example in the XML format.

  In the service information 420B of FIG. 11, <id_kind>, <groups>, and <locations> are defined as main tags. <Group> is defined as the first sub tag of <groups>. Also, <location> is defined as the first sub tag of <locations>, and the second sub tag of <name> and <address> is defined in <location>.

  <Id_kind> indicates the type of ID. Here, a case where the service provided by the content browsing application of the present embodiment is RSS will be described as an example. In this case, <id_kind> in the service information 420B is fixed as an ID indicating the type as RSS.

<Groups> indicates a group set in RSS. Further, <group> as the first sub tag of <groups> indicates an individual group in <groups>.
In the present embodiment, a plurality of RSSs having different contents depending on each predetermined place are set. <Locations> indicates locations set in RSS. <Location> which is the first sub tag of <locations> individually indicates a location for each RSS. <Name> indicates the name of the location indicated by <location>. <Address> indicates the address of the location indicated by <location>.

[Taste analysis information database]
Next, a structural example of the preference analysis information database 221 stored in the server storage unit 220 of the preference analysis server 204 will be described. As shown in FIG. 4, the preference analysis information database 221 includes a user information table 510, a terminal state history table 520, a search word table 530, and a content browsing status history table 540.

FIG. 12 shows a structural example of the user information table 510. Note that the user information table 510 shown in this figure has a structure corresponding to one user.
The user information table 510 illustrated in FIG. 12 includes columns of UserID 510a, UserName 510b, and modCount 510c.

UserID 510a indicates a user ID that uniquely identifies the corresponding user.
UserName 510b indicates the user name of the corresponding user.
modCount 510c is a counter indicating the timing for updating the recommended content list. Each time the terminal state history management unit 210 receives the terminal state information 410 for the corresponding user and stores it in the terminal state history table 520, the terminal state history management unit 210 decrements modCount 510c.
A state where the value of modCount 510c is “0” indicates that the timing for updating the recommended content list has been reached. The recommended content determination unit 230 executes control for newly generating a recommended content list in response to the value of modCount 510c being decremented to “0”.
Further, after the modCount 510c becomes “0”, the terminal state history management unit 210 resets the modCount 510c to a predetermined maximum value (for example, 100), and then stores the terminal state information 410 in the terminal state history table 520. Decrement each time it is stored.

FIG. 13 shows an example of the structure of the terminal state history table 520. Note that the terminal state history table 520 shown in this figure has a structure corresponding to one user.
The terminal state history table 520 illustrated in Fig. 13 includes Dbid 520a, UserID 520b, Tap_vertical_top 520c, Tap_vertical_middle 520d, Tap_vertical_bottom 520e, Tap_horizontal_left 520f, Tap_horizonatal_center 520g, Tap_horizontal_right 520h, use_time20

Dbid 520a indicates the ID of the terminal state history table 520.
UserID 520b indicates the user ID of the corresponding user.

Tap_vertical_top 520c indicates the number of times that a tap operation has been performed on the upper region (upper row) in the vertical (vertical) direction of the screen of the content browsing application.
Tap_vertical_middle 520d indicates the number of times that a tap operation has been performed on the central area (middle stage) in the vertical direction of the screen of the content browsing application.
Tap_vertical_bottom 520e indicates the number of times that a tap operation has been performed on the lower region (lower row) in the vertical direction of the screen of the content browsing application.

Tap_horizontal_left 520f indicates the number of times that a tap operation has been performed on the left region in the horizontal (horizontal) direction of the screen of the content browsing application.
Tap_horizontal_center 520g indicates the number of times that a tap operation has been performed on the center area in the horizontal direction of the screen of the content browsing application.
Tap_horizontal_right 520h indicates the number of tap operations performed on the right region in the horizontal direction of the screen of the content browsing application.

  As described above, Tap_vertical_top 520c, Tap_vertical_middle 520d, Tap_vertical_bottom 520e, Tap_horizontal_left 520f, Tap_horizontal_center 520g, and Tap_horizontal_right 520h are operation position history information indicating the history of positions where operations have been performed on the screen.

use_timezone_morning 520i indicates the number of times that a tap operation has been performed on the screen of the content browsing application in the morning time zone (for example, from 5 am to 11 am). Note that the tap operation here may be for any of the service list area AR1, the content list area AR2, and the download list area AR3.
use_timezone_daytime 520j indicates the number of times that a tap operation has been performed on the screen of the content browsing application in a daytime period (for example, 11:00 am to 6:00 pm).
use_timezone_night 520k indicates the number of times that a tap operation has been performed on the screen of the content browsing application in a night time zone (for example, from 6:00 pm to midnight).
use_timezone_midnight 520l indicates the number of times that a tap operation has been performed on the screen of the content browsing application in the time zone as midnight (for example, from 0:00 am to 5:00 am).

  As described above, use_timezone_morning 520i, use_timezone_daytime 520j, use_timezone_night 520k, and use_timezone_midnight 520l are time zone history information indicating the history of the time zone in which the user performed the operation on the terminal device 100.

With reference to FIGS. 14 to 16, an example of region division for the screen of the content browsing application corresponding to the operation position history information in the terminal state history table 520 will be described.
FIG. 14A shows a screen of the content browsing application displayed on the display unit 105. First, for the upper area corresponding to Tap_vertical_top 520c, a service list area AR1 and an area in which the thumbnail display positions Dpos2 whose display position numbers are 1 to 6 in the content list area AR2 are allocated.
For the middle area corresponding to Tap_vertical_middle 520d, the service list area AR1 and the area where the first to sixth thumbnail display positions Dpos2 are arranged in the content list area AR2 are allocated.
A download list area AR3 is allocated to the lower area corresponding to Tap_vertical_bottom 520e.

  Further, when the download list area AR3 is not displayed as shown in FIG. 14B, the thumbnail display with the display position numbers 13 to 18 in the content list area AR2 is displayed for the lower area on the screen of the content browsing application. An area in which the position Dpos2 is arranged is set.

In the content list area AR2, the left area corresponding to Tap_horizontal_left 520f is the left column of the thumbnail display positions Dpos2 arranged in three columns as shown in FIGS. 14 (a) and 14 (b). The area including the thumbnail display position Dpos2 (display position numbers 1, 4, 7,...) Is set.
In the content list area AR2, the left area corresponding to Tap_horizontal_center 520g is the thumbnail display position Dpos2 (display position numbers 2, 5, 8,...) In the middle column among the thumbnail display positions Dpos2 arranged in three columns. -Set the area including.
Also, in the content list area AR2, the left area corresponding to Tap_horizontal_right 520h, among the thumbnail display positions Dpos2 arranged in three columns, the thumbnail display position Dpos2 (display position numbers 6, 9, 12,... -Set the area including.

FIG. 15 also shows a service list area AR1 on the screen of the content browsing application. In the present embodiment, the service list area AR1 is divided into five thumbnail display positions Dpos1 in the horizontal (lateral) direction as shown in the figure. In this case, for the left area corresponding to Tap_horizontal_left 520f, the area of the thumbnail display position Dpos2 whose display position numbers are 1 and 2 is set.
For the central area corresponding to Tap_horizontal_center 520g, the area of the thumbnail display position Dpos2 whose display position number is 3 is set.
For the right area corresponding to Tap_horizontal_right 520h, the area of the thumbnail display position Dpos2 with display position numbers 4 and 5 is set.

FIG. 16 shows a download list area AR3 on the screen of the content browsing application.
Thus, when five thumbnail display positions Dpos3 are arranged in the horizontal direction with respect to the download list area AR3, the area of the thumbnail display position Dpos3 whose display position numbers are 1 and 2 for the left area corresponding to Tap_horizontal_left 520f. Set.
For the central area corresponding to Tap_horizontal_center 520g, the area of the thumbnail display position Dpos3 with the display position number 3 is set.
For the right area corresponding to Tap_horizontal_right 520h, the area of the thumbnail display position Dpos3 whose display position numbers are 4 and 5 is set.

The terminal state information processing unit 102 updates the operation position history information in the terminal state history table 520 based on the received list_kind 416 and disp_pos 417 in the terminal state information 410.
That is, the terminal state information processing unit 102 refers to the list_kind 416 and recognizes which of the service list area AR1, the content list area AR2, and the download list area AR3 has been tapped. At this time, if it is recognized that the tap operation has been performed on the service list area AR1, the terminal state information processing unit 102 recognizes that the vertical area where the tap operation has been performed is the upper area. If it is recognized that a tap operation has been performed on the download list area AR3, the terminal state information processing unit 102 recognizes that the vertical area where the tap operation has been performed is the lower area.

In addition, when the tapped area is the content list area AR2, the terminal state information processing unit 102 further refers to disp_pos 417 and displays the display position number of the thumbnail display position Dpos2 where the tapping operation is performed in the content list area AR2. Recognize
If the recognized display position number is any one of Nos. 1 to 6, the terminal state information processing unit 102 recognizes that the vertical region where the tap operation has been performed is the upper region.
Further, if the recognized display position number is any one of Nos. 7 to 12, the terminal state information processing unit 102 recognizes that the vertical area where the tap operation is performed is the middle area.
Further, if the recognized display position number is any one of Nos. 13 to 18, the terminal state information processing unit 102 recognizes that the vertical area where the tap operation is performed is the lower area.

Further, the terminal state information processing unit 102 recognizes the horizontal direction as follows.
First, when the tapped area is the content list area AR2, the terminal state information processing unit 102 refers to disp_pos 417 and sets the display position number of the thumbnail display position Dpos2 where the tapping operation is performed in the content list area AR2. recognize.
If the recognized display position number corresponds to one of the thumbnail display positions Dpos2 included in the left column, the terminal state information processing unit 102 indicates that the vertical area where the tap operation has been performed is the left area. Recognize.
Further, if the recognized display position number corresponds to any one of the thumbnail display positions Dpos1 included in the center column, the terminal state information processing unit 102 determines that the vertical area where the tap operation has been performed is the center area. Recognize that
Further, if the recognized display position number corresponds to one of the thumbnail display positions Dpos1 included in the right column, the terminal state information processing unit 102 determines that the vertical area where the tap operation has been performed is the right area. Recognize that

When the tapped area is the service list area AR1, the terminal state information processing unit 102 refers to disp_pos 417 and sets the display position number of the thumbnail display position Dpos1 where the tapping operation is performed in the service list area AR1. recognize.
If the recognized display position number is 1 or 2, the terminal state information processing unit 102 recognizes that the vertical region where the tap operation has been performed is the left region.
Further, if the recognized display position number is 3, the terminal state information processing unit 102 recognizes that the vertical region where the tap operation has been performed is the central region.
Further, if the recognized display position number is 4 or 5, the terminal state information processing unit 102 recognizes that the vertical region where the tap operation has been performed is the right region.
Even when the tapped area is the download list area AR3, the terminal state information processing unit 102 is tapped in the horizontal direction by the same processing as when the tapped area is the service list area AR1. It can be recognized whether the area is left, center, or right.

FIG. 17 shows an example of the structure of the search term table 530. The search term table 530 shown in this figure has a structure corresponding to one user.
The search word table 530 includes Dbid 530a, UserID 530b, data_1st 530c, data_2nd 530d, data_3rd 530e, data_4th 530f, data_5th 530g, rensou_keyword_1st 530h, rensou_keyword_2nd 530i, rensou_keyword_3rd530j, rensth_key30

Dbid 530a indicates the ID of the search term table 530.
UserID 530b indicates the user ID of the corresponding user.
data_1st 530c is a first associative word search word.
data_2nd 530d is the second search term for associative words.
data_3rd 530e is a third associative word search word.
data_4th 530f is a fourth associative word search word.
data_5th 530g is a fifth associative word search word.
rensou_keyword_1st 530h is a first recommended content search word.
rensou_keyword_2nd 530i is a second recommended content search word.
rensou_keyword_3rd 530j is a third recommended content search term.
rensou_keyword_4th 530k is a fourth recommended content search term.
rensou_keyword_5th 530l is a fifth recommended content search term.

FIG. 18 shows a structure example of the content browsing status history table 540. The content browsing status history table 540 shown in this figure has a structure corresponding to one user.
The content browsing status history table 540 corresponding to one user includes one or more individual content correspondence tables 541 (540-1, 540-2, 540-3,...).

One individual content correspondence table 541 is a table showing a browsing situation corresponding to a browsing event of one content.
The individual content correspondence table 541 includes columns of Dbid 541a, UserID 541b, dataID 541c, Kind 541d, Tap_count 541e, read_average_time 541f, read_total_count 541g and read_total_time 541h, respectively.

Dbid 541a indicates the ID of the individual content correspondence table 541.
UserID 541b indicates the user ID of the corresponding user.
The data ID 541c indicates the ID (content_id or service_id) of the corresponding content or service.
Kind 541d indicates the type of content data (DATA) as a book or RSS as a service.

Tap_count 541e indicates the number of times that a tap operation has been performed during browsing display (view display or preview display) of the corresponding content.
The read_average_time 541f indicates an average value (viewing average time) of browsing time per browsing event for the corresponding content. A browsing event is an event that occurs in response to an operation for browsing, for example, and makes the content available for browsing. This browsing event ends, for example, in response to an operation that ends browsing the content.
The read_total_count 541g indicates the total number of times that a browsing event has occurred for the corresponding content (total number of browsing times).

read_total_time 541h indicates the total browsing time for the corresponding content. The total browsing time indicated by this read_total_time 541h is obtained using the parameters of event_start_date 412 and event_end_date 413 in the terminal state information 410 transmitted by the terminal device 100 in response to the browsing operation for the corresponding content.
That is, each time the terminal state history management unit 210 receives the terminal state information 410 for the corresponding content, the terminal state history management unit 210 subtracts the event_start_date 412 from the event_end_date 413 to obtain a viewing time corresponding to one event. The terminal state history management unit 210 updates the read_total_time 541h by adding the browse time thus obtained to the value of the read_total_time 541h.

[Example of method for determining search words for associative words]
The associative word search word determination unit 240 in the preference analysis server 204 determines first to fifth associative word search words based on the received terminal state information 410 when transmitting the search request in step S5 of FIG. . The determined search words for first to fifth associative words are stored as values of data_1st 530c, data_2nd 530d, data_3rd 530e, data_4th 530f, and data_5th 530g in the search word table 530, respectively.
The associative word search word determination unit 240 transmits an associative word search request corresponding to the first to fifth associative word search words to the associative word search server 205.

The search word table 530 at the stage where the first to fifth associative word search words have not been determined is, for example, as shown in FIG. Not stored.
Also, the recommended content search word is determined using the associated word searched using the first to fifth associated word search words as search keys. For this reason, the rensou_keyword_1st 530h, rensou_keyword_2nd 530i, rensou_keyword_3rd 530j, rensou_keyword_4th 530k, and rensou_keyword_5th 530l in the search word table 530 are not stored with valid values at this stage.

In the present embodiment, the first to fifth search words for associative words are defined as follows.
The search word for the first associative word is a noun having the highest importance among nouns in the morpheme acquired by performing morphological analysis on the character string acquired from the content information of the content having the largest number of tap operations. .
The second associative word search word is a noun having the highest importance among the nouns in the morpheme acquired by performing the morphological analysis on the character string acquired from the content information 420A of the content having the longest browsing time.
The third associative word search term is a noun having the highest importance among nouns as morphemes obtained by morphological analysis of the character string obtained from the service information 420B of the service having the largest number of tap operations. It is.
The fourth associative word search word was tapped in the upper, middle, lower, left, right, and center areas of the content browsing application screen (see FIGS. 14 to 16). This is a phrase obtained by converting the most frequently used area into a character string. As an example, the phrases as the search words for the fourth associative words are “upper”, “middle”, “lower” corresponding to the upper, middle, lower, left, right, and center of the screen of the content browsing application. , “Left”, “Right”, “Middle”. That is, the phrase as the fourth associative word search word is any one of “upper”, “middle”, “lower”, “left”, and “right”.
The fifth associative word search word is a phrase obtained by converting a time zone in which an event occurs most frequently into a character string. As an example, the fifth associative word search word is one of morning, noon, night, and midnight.

The associative word search word determination unit 240 determines the first associative word search word as follows.
In the associative word search word determination unit 240, first, the content selection unit 241 refers to the content browsing status history table 540 of FIG. Select the most content.
Of the individual content correspondence table 541 in the content browsing status history table 540, the one that stores information about the content includes the character string “DATA” in the data_ID 541c. Further, the number of times the tap operation has been performed is stored in Tap_count 541e in the individual content correspondence table 541.

  Therefore, the content selection unit 241 selects the individual content correspondence table 541 having the maximum value stored in the Tap_count 541e from the individual content correspondence table 541 including the character string “DATA” in the data_ID 541c. The value stored in the data_ID 541c of the individual content correspondence table 541 selected in this way is the content_ID of the content with the most tap operations. The content selection unit 241 acquires content_ID stored in data_ID 541c of the selected individual content correspondence table 541 as a selection result of the content with the largest number of tap operations.

The content selection unit 241 transmits a content information request to the content distribution server 201 together with the content_ID acquired as a content selection result in step S3 of FIG.
The content selection unit 241 receives the content information 420A transmitted from the content distribution server 201 in response to the content information request in step S4. The content information 420A received in this way includes a character string as illustrated in FIG.

The morpheme analysis unit 242 extracts a character string from the content information 420A, and performs morpheme analysis on the extracted character string.
Next, the morpheme correspondence determining unit 243 determines the first search word for associative words from the morphemes obtained by the morpheme analysis as follows.
That is, the morpheme correspondence determining unit 243 obtains the importance for each noun among the morphemes obtained by the morpheme analysis. The method for obtaining the importance is not particularly limited, but the importance can be set based on the appearance frequency, for example. The importance can also be set based on the connection information obtained by counting the types of words to be connected. It can also be set based on information indicating how many single nouns can be connected in terms of information theory called index perplexity. Moreover, you may set importance based on 2 or more information of these appearance frequency, connection information, and parameter | index perplexity. The morpheme correspondence determining unit 243 determines the noun having the highest importance as the first associative word search word.
The morpheme correspondence determining unit 243 stores the determined first associative word search word in data_1st 530c of the search word table 530, for example, as shown in FIG. FIG. 19 shows an example in which the word / phrase determined as the first associative word search word is “Company A”.

The associative word search word determination unit 240 determines the second associative word search word as follows.
In the associative word search word determination unit 240, first, the content selection unit 241 refers to the content browsing status history table 540 of FIG. 18 and sets the browsing time as the content to be used for the determination of the associative word search word. Select the longest content.
As described above, in the content browsing status history table 540, the individual content correspondence table 541 that stores information about content includes the character string “DATA” in the data_ID 541c.
The content browsing time is stored as the total browsing time in read_total_time 541h in the individual content correspondence table 541.

  Therefore, the content selection unit 241 selects the individual content correspondence table 541 having the maximum value stored in the read_total_time 541h from the individual content correspondence table 541 including the character string “DATA” in the data_ID 541c. The content selection unit 241 acquires content_ID stored in data_ID 541c of the selected individual content correspondence table 541 as a selection result of the content having the longest browsing time.

The content selection unit 241 transmits a content information request to the content distribution server 201 together with the content_ID acquired as a result of content selection in step S3 of FIG.
The content selection unit 241 receives the content information 420A transmitted from the content distribution server 201 in response to the content information request in step S4.

The morpheme analysis unit 242 extracts a character string as an element item from the content information 420A, and executes morpheme analysis on the extracted character string.
The morpheme correspondence determining unit 243 determines the second associative word search word from the morphemes obtained by the morpheme analysis, for example, in the same procedure as when determining the first associative word search word.
The morpheme correspondence determining unit 243 stores the determined second associative word search word in data_2nd 530d of the search word table 530, for example, as shown in FIG. FIG. 19 shows an example in which the phrase determined as the second associative word search word is “ware”.

The associative word search word determination unit 240 determines the third associative word search word as follows.
In the associative word search word determination unit 240, first, the content selection unit 241 refers to the content browsing status history table 540 of FIG. Select the service with the highest number of times.
As described above, in the content browsing status history table 540, the individual content correspondence table 541 that stores information about the service includes the character string “RSS” in the data_ID 541c. The number of tap operations for the service is stored in Tap_count 541e in the individual content correspondence table 541.

  The content selection unit 241 selects the individual content correspondence table 541 having the maximum value stored in the Tap_count 541e from the individual content correspondence table 541 including the character string “RSS” in the data_ID 541c. The content selection unit 241 acquires service_ID stored in the data_ID 541c of the selected individual content correspondence table 541 as a selection result of the content with the longest browsing time.

The content selection unit 241 transmits a content information request (service information request) to the content distribution server 201 together with service_ID acquired as a content selection result in step S3 of FIG.
The content selection unit 241 receives the service information 420B transmitted from the content distribution server 201 in response to the service information request in step S4.

The morpheme analysis unit 242 extracts a character string as an element item from the service information 420B, and executes morpheme analysis on the extracted character string.
The morpheme correspondence determination unit 243 determines the third associative word search word from the morphemes obtained by the morpheme analysis, for example, in the same procedure as when determining the first associative word search word.
The morpheme correspondence determining unit 243 stores the determined third associative word search word in data_3rd 530e of the search word table 530 as shown in FIG. FIG. 19 shows an example in which the word / phrase determined as the third associative word search word is “Kanagawa”.

The associative word search word determination unit 240 determines the fourth associative word search word as follows.
In the associative word search word determination unit 240, the operation position correspondence determination unit 244 refers to the operation position history information in the terminal state history table 520.
The operation position history information includes columns of Tap_vertical_top 520c, Tap_vertical_middle 520d, Tap_vertical_bottom 520e, Tap_horizontal_left 520f, Tap_horizonatal_center 520g, and Tap_horizontal_right 520h.

Each of the operation position history information is associated with an associated word search word in advance as follows.
The word “upper” is associated with Tap_vertical_top 520c.
The word “medium” is associated with Tap_vertical_middle 520d.
The word “below” is associated with Tap_vertical_bottom 520e.
The word “left” is associated with Tap_horizontal_left 520f.
The word “medium” is associated with Tap_horizonatal_center 520g.
The word “right” is associated with Tap_horizontal_right 520h.

The operation position correspondence determining unit 244 identifies the operation position history information having the maximum value from the operation position history information. The operation position correspondence determining unit 244 determines a word associated with the specified operation position history information as the fourth associative word search word.
The operation position correspondence determining unit 244 stores the determined fourth associative word search word in data_4th 530f of the search word table 530 as shown in FIG. FIG. 19 shows an example in which “left” determined as the fourth associative word search word is stored in accordance with the maximum value stored in Tap_horizontal_left 520f.

The associative word search word determination unit 240 determines the fifth associative word search word as follows.
In the associative word search word determination unit 240, the operation time correspondence determination unit 245 refers to the time zone history information in the terminal state history table 520.
The time zone history information is each column of use_timezone_morning 520i, use_timezone_daytime 520j, use_timezone_night 520k, and use_timezone_midnight 520l.

Each time zone history information is associated with words in advance as follows.
The word “morning” is associated with use_timezone_morning 520i.
The term “daytime” is associated with use_timezone_daytime 520j.
The word “night” is associated with use_timezone_night 520k.
Use_timezone_midnight520l is associated with the phrase “midnight”.

The operation time correspondence determining unit 245 identifies the time zone history information having the maximum value from the time zone history information. The operation time correspondence determining unit 245 determines the word / phrase associated with the specified operation position history information as the fourth associative word search word.
The operation time correspondence determining unit 245 stores the determined fifth associative word search word in data_5th 530g of the search word table 530 as shown in FIG. FIG. 19 shows an example in which “night” determined as the fifth associative word search word is stored in accordance with the maximum value stored in use_timezone_night 520k.

[Associative word search example]
The morpheme correspondence determining unit 243 generates an associative word search request for each of the first to third associative word search words stored in data_1st 530c, data_2nd 530d, and data_3rd 530e as shown in FIG. 19 as step S5 in FIG. To 205.
Further, the operation position correspondence determination unit 244 transmits an associative word search request for the fourth associative word search word stored in data_4th 530f as shown in FIG.
Also, the operation time correspondence determining unit 245 transmits an associative word search request for the fifth associative word search word stored in data_5th 530g as shown in FIG. 19 to step S5 in FIG.

The associative word search unit 250 in the associative word search server 205 searches for the associative word corresponding to each of the first to fifth associative word search words in response to each of the associative word search requests. As an example, the associative word search unit 250 outputs a predetermined number (for example, about 50) of associative words as a search result corresponding to the first associative word search word. Similarly, the associative word search unit 250 outputs a predetermined number of associative words as search results corresponding to the second to fifth associative word search words.
Then, the associative word search unit 250 transmits search results corresponding to the first to fifth associative word search words to the preference analysis server 204 in step S6 of FIG.

[Example of method for determining search terms for recommended content]
Next, an example of a method for the recommended content search word determination unit 260 to determine the recommended content search word will be described.
The recommended content search word determination unit 260 inputs the search results for each of the first to fifth associative word search words transmitted from the associative word search server 205 as described above. The recommended content search word determination unit 260 determines the first to fifth recommended content search words as follows using the input search results for the first to fifth associative word search words. To do.

First, the recommended content search word determination unit 260 determines the first recommended content search word as follows.
The recommended content search word determination unit 260 randomly selects a predetermined number of associative words from the associative words that are search results corresponding to the first associative word search word. As a specific example, the recommended content search word determination unit 260 randomly selects five associative words from 50 associative words that are search results corresponding to the first associative word search word. Next, the recommended content search word determination unit 260 determines, as a first recommended content search word, a shortest character string from among randomly selected associative words.
The recommended content search word determination unit 260 stores the first recommended content search word determined as described above in the rensou_keyword_1st 530h of the search word table 530, as shown in FIG.

  The recommended content search word determination unit 260 determines the remaining second to fifth recommended content search words in the same manner as the first recommended content search word. That is, the recommended content search word determination unit 260 randomly selects a predetermined number of associative words from among the associative words corresponding to the second to fifth associative word search words.

The recommended content search word determination unit 260 selects the shortest character string in each of the associative words randomly selected from the search results corresponding to the second to fifth associative word search words. It is determined as the second to fifth recommended content search terms.
As shown in FIG. 20, the recommended content search word determination unit 260 determines the rensou_keyword_2nd 530i, rensou_keyword_3rd 530j, rensou_keyword_4th 530k, and rensou_keyword_5th 530l of the search word table 530 as shown in FIG. Store in each.

  In the present embodiment, among the first to fifth associative word search terms, the first to third associative word search terms are determined based on content that matches the user's preference. On the other hand, the search words for the fourth and fifth associative words are determined not based on the content but based on the position and time of the operation performed on the screen of the content browsing application by the user. And in this embodiment, an associative word is searched combining these search words for associative words. Thereby, in this embodiment, for example, a search result different from the case of searching for an associative word using only the associative word search word determined based on the content that matches the user's preference is obtained.

Here, FIG. 21A shows the result of searching for an associative word only by “yugawara”, which is an associative word search word determined based on content that matches the user's preference. On the other hand, FIG. 21B is a combination of the above-mentioned “yugawara” associative word search word and the “upper” associative word search word determined based on the operation position on the screen of the content browsing application. The result of searching the associative word with the search term “Yugaharakami” is shown.
As can be seen by comparing FIG. 21 (a) and FIG. 21 (b), there is a difference in the search result of the associated words.

As a result of the search for associated words differing as described above, there is a difference in the determination result of the search word for the recommended content list between the case of FIG. 21A and the case of FIG. There are also differences in the contents of the list.
As a result, in this embodiment, the content of the recommended content list is made surprising by combining the search word for associative words based on the content that matches the user's preference with the search word for associative words based on the operation result for the screen. Can be given.

[Example of recommended content list acquisition method]
Next, an example of a method for obtaining a recommended content list will be described.
The recommended content search word determination unit 260 transmits a recommended content list request to the content management server 202 together with the first to fifth recommended content search words determined as described above, as step S7 of FIG.
In the content management server 202, the recommended content search unit 270 determines recommended content to be included in the recommended content list as described below in response to reception of the recommended content list request.

  The recommended content search unit 270 searches the content meta information storage unit 271 for content including at least one of the first to fifth recommended content search terms received together with the recommended content list request in the meta information. The content searched by the recommended content search unit 270 is the recommended content.

FIG. 22 shows the content of information stored in the content meta information storage unit 271.
The content meta information storage unit 271 stores a book entry table 610 and a book information table 620.
The book entry table 610 is an entry table corresponding to each registered electronic book. The book information table 620 is a table for storing meta information based on predetermined attributes for registered electronic books.

  Since the content management server 202 manages the electronic books registered for each user in units of bookshelves, the content meta information storage unit 271 also stores a table for managing bookshelves. Since the table for managing the bookshelf need not be used for determining recommended content, the illustration thereof is omitted in FIG.

FIG. 23 shows an example of the structure of the book entry table 610.
The book entry table 610 has a structure including id 610a, comment 610b, score 610c, categories 610d, modtime 610e, clicktime 610f, and book_id 610g.
The id 610a is an ID that uniquely identifies the book entry table 610.
Comment 610b is comment information. The comment information is information indicating a comment about the corresponding electronic book by a character string (for example, text data).
The score 610c is score information and indicates an evaluation value for the corresponding electronic book.
Categories 610d is field information, and indicates predetermined information related to the book such as the date of issue, in addition to the field of the corresponding electronic book.
modtime 610e is update information and indicates the date and time when the book entry table 610 was updated.
book_id 610g is a book information table ID and indicates the ID of the book information table corresponding to the corresponding book.

FIG. 24 shows a structural example of the book information table 620.
The book information table 620 has a structure including id 620a, isbn 620b, title 620c, publisher 620d, authors 620e, price 620f, imageurl 620g, and modtime 620h.
The id 620a is a book information table ID and indicates an ID for uniquely identifying the book information table 620. The book entry table 610 and the book information table 620 for the same electronic book are associated by the book information table ID indicated by the id 620a and the book information table ID indicated by the book_id 610g.
isbn 620b is ISBN (International Standard Book Number) information and indicates the ISBN assigned to the corresponding electronic book.
The title 620c is title information and indicates the title of the corresponding electronic book by a character string.
Publisher 620d is publisher information, and indicates the publisher of the corresponding electronic book by a character string.
Authors 620e is author information and indicates the author of the corresponding electronic book by a character string.
The price 620f is price information and indicates the price of the corresponding electronic book by a numerical character string.
The imageurl 620g indicates a URL (Uniform Resource Locator) where thumbnail image data for a corresponding electronic book is stored.
modtime 620h is update information, and indicates the date and time when the book information table 620 was updated.

The recommended content search unit 270 searches the book entry table 610 and the book information table 620 including character strings that match any of the first to fifth recommended content search terms received together with the recommended content list request. At this time, the recommended content search unit 270 may search for any column in the book entry table 610 and the book information table 620 without any particular limitation.
The recommended content search unit 270 determines the book corresponding to the book entry table 610 or the book information table 620 searched as described above as the recommended content. Then, the recommended content search unit 270 generates a recommended content list that is a list of content determined as recommended content.

FIG. 25 shows a structural example of the recommended content list 700. The recommended content list 700 shown in this figure is, for example, in XML format.
In the recommended content list 700, <book> is defined as a main tag in the <entry> tag. The main tag of <book> describes meta information about one electronic book. That is, the recommended content list 700 has a structure in which meta information for each electronic book determined as recommended content is described in <entry> tags in units of main tags of <book>.

In the recommended content list 700, <title>, <publisher>, <authors>, <imageurl>, <comment>, and <score> are defined as sub tags in the main tag of <book>.
The subtitle of <title> is title information, and the contents of title 620c in the book information table 620 are described.
A sub tag of <publisher> is publisher information, and describes the contents of the publisher 620d in the book information table 620.
<Authors> is author information and describes the contents of authors 620e in the book information table 620.
A sub tag of <imageurl> is a URL storing thumbnail images, and describes the contents of imageurl 620g in the book information table 620.
The sub tag of <comment> is comment information, and describes the content of the comment 610b in the book entry table 610.
The <score> tag is score information and describes the contents of the score 610c in the book entry table 610.

The recommended content search unit 270 transmits the recommended content list 700 generated as described above to the preference analysis server 204 in step S8 of FIG. 2 together with the user_ID of the corresponding user.
The recommended content search word determination unit 260 in the preference analysis server 204 transfers the received recommended content list 700 and user_ID to the content distribution server 201 in step S9 of FIG.
In the content distribution server 201, the recommended content list distribution unit 281 holds the received recommended content list 700. Then, for example, in response to a recommended content list request being transmitted from the terminal device 100, the request is transmitted to the terminal device 100 of the user of User_ID received together with the recommended content list request in step S10 of FIG.

[Example of processing procedure]
The flowchart in FIG. 26 illustrates an example of a processing procedure executed by the terminal device 100, the content distribution server 201, and the preference analysis server 204 in the content recommendation system of the present embodiment for content recommendation.
In the terminal device 100, the terminal state information processing unit 102 waits for a state change to occur in the terminal device 100 in accordance with the use of the user (procedure for returning to step S101 in steps S101-NO and S104-NO). In correspondence with the description so far, a change in the state of the terminal device 100 according to the use of the user corresponds to an operation being performed on the screen of the content browsing application and an event being generated accordingly. .

  In response to a state change occurring in the terminal device 100 according to the user's use (step S101-YES), the terminal state information processing unit 102 generates terminal state information 410 corresponding to the state change (operation) (step). S102). Next, the terminal state information processing unit 102 holds the generated terminal state information 410 (step S103). At this time, the terminal state information processing unit 102 may store the terminal state information 410 in the terminal storage unit 104, for example.

Next, the terminal state information processing unit 102 determines whether or not the terminal state information transmission timing described with reference to FIG. 2 has been reached (step S104). When it is not a transmission timing (step S104-NO), the terminal state information processing part 102 returns to step S101.
On the other hand, when the transmission timing is reached (step S104-YES), the terminal state information processing unit 102 transmits the terminal state information held so far to the preference analysis server 204 (step S105), and returns to step S101. .

In the preference analysis server 204, the terminal state history management unit 210 waits for the terminal state information 410 to be received (step S201—NO).
In response to the reception of the terminal state information 410 (step S201—YES), the terminal state history management unit 210 registers the received terminal state information 410 in the terminal state history table 520 of the server storage unit 220 (step S201). S202). The terminal state history table 520 decrements the value of modCount 510c in the user information table 510 of the user of the terminal device 100 that is the transmission source of the terminal state information 410, according to the number of registered terminal state information 410.

    Next, the recommended content determination unit 230 determines whether or not it is time to generate a recommended content list (step S203). For this reason, the recommended content determination unit 230 determines whether or not the value of the modCount 510c is “0”.

If it is not time to generate the recommended content list (step S203—NO), the terminal state history management unit 210 returns the process to step S101.
On the other hand, when it is time to generate the recommended content list, the recommended content determination unit 230 executes a process for generating the recommended content list (step S204). In generating the recommended content list, the recommended content determination unit 230 cooperates with the content distribution server 201, the content management server 202, the associative word search server 205, and the like as described above.
The recommended content determination unit 230 transfers the generated recommended content list to the content distribution server 201 (step S205).

In the content distribution server 201, the recommended content list distribution unit 281 waits for the reception of the recommended content list transmitted from the preference analysis server 204 (step S301-NO).
In response to receiving the recommended content list (step S301—YES), the recommended content list distribution unit 281 stores the received recommended content list.
Note that the recommended content list stored in this manner is transmitted to the requesting terminal device 100 in response to a request from the terminal device 100.

  FIG. 27 shows a procedure example of a recommended content list generation process executed by the recommended content determination unit 230 of the preference analysis server 204 as step S204 of FIG. Along with this, FIG. 27 shows a procedure example of processing executed by the content distribution server 201, the content management server 202, and the associative word search server 205 in accordance with the recommended content list generation processing of the recommended content determination unit 230.

  In the preference analysis server 204, the content selection unit 241 of the recommended content determination unit 230 reads the terminal state history table 520 corresponding to the target user from the server storage unit 220 (step S401).

  Next, the content selection unit 241 selects a content and a service to be used for determining an associative word search word (step S402). For example, as described above, the content selection unit 241 selects the content with the largest number of tap operations, the content with the longest browsing time, and the service with the largest number of tap operations.

  The content selection unit 241 transmits a content information request for requesting the content information 420A and service information 420B corresponding to the content and service selected in step S402 to the content distribution server 201 (step S403).

  The morpheme analysis unit 242 performs morpheme analysis on the content information and service information transmitted from the content distribution server 201 in response to the content information request transmitted in step S403 (step S404).

Next, the associative word search word determination unit 240 determines, for example, an associative word search word (step S405).
In step S405, the morpheme correspondence determining unit 243 determines first to third associative word search words from the morphemes obtained as a result of the morpheme analysis in step S404, for example, as described above.
In step S405, the operation position correspondence determination unit 244 determines the fourth search word for the associative word with reference to the operation position history information in the terminal state history table 520.
Further, as step S405, the operation time correspondence determining unit 245 determines the fifth associative word search word with reference to the time zone history information in the terminal state history table 520.
The associative word search word determination unit 240 transmits an associative word search request to the associative word search server 205 together with the associative word search word determined in step S405 (step S406).

  The recommended content search word determination unit 260 determines the recommended content search word using the associative word transmitted from the associative word search server 205 in response to the request in step S406 (step S407). For example, as described above, the recommended content search word determination unit 260 uses the first to fifth recommendation words using the associated words searched for the first to fifth associative word search words. Determine search terms for content.

The recommended content search word determination unit 260 transmits a recommended content list request to the content management server 202 together with the recommended content search word determined in step S407 (step S408).
The recommended content search word determination unit 260 transfers the recommended content list transmitted by the content management server 202 in response to the recommended content list request transmitted in step S408 to the content distribution server 201 (step S409).

  In the content distribution server 201, the content search unit 283 waits for reception of a content information request transmitted from the preference analysis server 204 in step S403 (step S501-NO).

  In response to receiving the content information request (step S501-YES), the content search unit 283 searches for the content or service of content_ID or service_ID received together with the content information request (step S502). Then, the content search unit 283 transmits the content information of the searched content and the content information of the searched content to the preference analysis server 204 (step S503).

  Also, the recommended content list distribution unit 281 in the content distribution server 201 stores the recommended content list transferred from the preference analysis server 204 in step S409 for distribution to the terminal device 100 (step S504). This step S504 is the same process as step S302 in FIG.

In the associative word search server 205, the associative word search unit 250 waits for reception of an associative word search request transmitted from the preference analysis server 204 in step S406 (step S601—NO).
In response to the associative word search request being received (step S601—YES), the associative word search unit 250 searches the associative word database 251 for the associative word corresponding to the associative word search word received together with the associative word search request. (Step S602). The associative word search unit 250 transmits the searched associative word to the preference analysis server 204 (step S603).

In the content management server 202, the recommended content search unit 270 waits for reception of a recommended content list request transmitted from the preference analysis server 204 in step S408 (NO in step S701).
When the recommended content list request is received (step S701—YES), the recommended content search unit 270 searches for recommended content using the recommended content search word received together with the recommended content list request (step S702). ). The recommended content search unit 270 generates a recommended content list 700 that lists the searched recommended content, and transmits the generated recommended content list 700 to the preference analysis server 204 (step S703).

[Modification]
In the description so far, the terminal state information 410 has contents as operation history information reflecting an operation result performed by the user on the screen of the content browsing application. Along with this, the terminal state history table 520 also shows an operation history performed by the user on the screen of the content browsing application.
The terminal state information 410 may include information indicating a predetermined state of the terminal device 100 in addition to the operation history information as described above.

  As an example, the terminal state information 410 may include information indicating activation of the content browsing application, information such as activation time, and end time. Accordingly, the terminal state history table 520 can include information such as the number of times the content browsing application is activated, the total activation time, and the activated time zone.

  Further, the terminal status information 410 may include a name and other meta information about the application installed in the terminal device 100. In response to this, the terminal state history table 520 includes information on the applications installed in the terminal device 100.

  Further, the terminal state information 410 may include a name and other meta information about a task (application) operating in parallel in the terminal device 100. In response to this, the terminal state history table 520 includes information on a task being operated in the terminal device 100.

  Further, information on detection outputs of various sensors provided in the terminal device 100 may be included in the terminal state information 410. Examples of such information include acceleration detected by an acceleration sensor, geomagnetism detected by a geomagnetic sensor, illuminance detected by an illuminance sensor, atmospheric pressure detected by an atmospheric pressure sensor, angular velocity detected by a gyro sensor, and current position detected by a positioning sensor Etc. In response to this, the terminal state history table 520 includes information on a task being operated in the terminal device 100.

Further, the terminal state information 410 may include information on whether or not the earphone is connected to the terminal device 100. In response to this, the terminal state history table 520 includes information indicating whether or not the earphone is connected in the terminal device 100.
Further, the connection status of a data interface such as USB (registered trademark) or Bluetooth (registered trademark) included in the terminal device 100 may be included in the terminal status information 410. In response to this, the terminal state history table 520 includes information indicating a connection state for each data interface in the terminal device 100.
Further, the terminal state information 410 may include radio wave intensity in wireless communication such as WiFi (registered trademark; Wireless fidelity). In response to this, the terminal state history table 520 includes information indicating the radio wave intensity for each wireless communication method.
Further, information on the remaining battery level for driving the terminal device 100 may be included in the terminal state information 410. In response to this, the terminal state history table 520 includes information indicating the remaining battery information.

  Further, information regarding the specifications of the display device as the display unit 105 may be included in the terminal state information 410. Information regarding the specifications of the display device is, for example, the screen size, dot density (dot per inch), and the like. In response to this, the terminal state history table 520 includes information indicating information on information regarding the specifications of the display device.

  Then, the associative word search word determination unit 240 determines the associative word search word by converting the information included in the terminal state history table 520 into a character string according to a predetermined rule, for example. Then, the associative word search server 205 searches for an associative word using, for example, the first to fifth associative word search words determined in the previous embodiment and the associative word search words determined as described above. To do. As a result, a further change can be obtained in the retrieval result of the associative word.

  Note that the configuration of the server system 200 shown in FIGS. 1 and 2 is merely an example. That is, a part of the servers constituting the server system 200 shown in FIGS. 1 and 2 may be integrated and realized as one server. Moreover, all the servers which comprise the server system 200 shown in FIG. 1 and FIG. 2 may be integrated and implement | achieved as one server.

  Further, in the description so far, the case where the electronic book is content is exemplified, but the present invention is not limited to this. For example, the configuration of the present invention is applicable to content such as advertisements and newspapers. Applicable.

  3 is recorded on a computer-readable recording medium, and the program recorded on the recording medium is read into a computer system and executed to recommend content. You may go. Here, the “computer system” includes an OS and hardware such as peripheral devices.

Further, the “computer system” includes a homepage providing environment (or display environment) if a WWW system is used.
The “computer-readable recording medium” refers to a storage device such as a flexible medium, a magneto-optical disk, a portable medium such as a ROM and a CD-ROM, and a hard disk incorporated in a computer system. Further, the “computer-readable recording medium” refers to a volatile memory (RAM) in a computer system that becomes a server or a client when a program is transmitted via a network such as the Internet or a communication line such as a telephone line. In addition, those holding programs for a certain period of time are also included. The program may be a program for realizing a part of the functions described above, and may be a program capable of realizing the functions described above in combination with a program already recorded in a computer system.

  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 designs and the like that do not depart from the gist of the present invention.

DESCRIPTION OF SYMBOLS 100 Terminal device 101 Operation input part 102 Terminal state information processing part 103 Recommended content list control part 104 Terminal memory | storage part 105 Display part 200 Server system 201 Content delivery server 202 Content management server 203 Terminal state information management server 204 Preference analysis server 205 Associative word Search server 210 Terminal state history management unit 220 Server storage unit 221 Preference analysis information database 230 Recommended content determination unit 240 Associative word search word determination unit 241 Content selection unit 242 Morphological analysis unit 243 Morphological correspondence determination unit 244 Operation position correspondence determination unit 245 Operation time correspondence determination unit 250 Associative word search unit 251 Associative word database 260 Recommended content search word determination unit 270 Recommended content search unit 271 Content meta information storage unit 281 Recommended content list distribution unit 282 Distribution content storage unit 283 Content search unit 300 Network

Claims (12)

It has a terminal device and a server that communicate with each other via a network,
The terminal device
An operation input unit for inputting an operation;
A terminal state information processing unit that generates terminal state information including the content of the operation input by the operation input unit as information indicating a predetermined state of the terminal device, and transmits the generated terminal state information to the server;
A recommended content list control unit that executes at least control for displaying a recommended content list received from the server on a display unit;
The server
A terminal state history management unit for registering terminal state information transmitted from the terminal device in a terminal state history;
A server storage unit for storing the terminal state history;
Based on the terminal state history, recommended content to be recommended to the user of the terminal device is determined from recommended candidate content, and the recommended content list indicating the determined recommended content is transmitted to the terminal device A content recommendation system. The recommended content determination unit includes:
An associative word search word determining unit that determines an associative word search word that is a character string for searching for an associative word based on the terminal state history;
An associative word search unit for searching for an associative word based on the associative word search word;
A recommended content search word determination unit that determines a recommended content search word that is a character string for searching for recommended content based on the associative word searched by the associative word search unit;
A recommended content search unit for determining, as the recommended content, content searched from recommended candidate content based on the recommended content search terms;
The content recommendation system according to claim 1. The associative word search term determination unit
A content selection unit that selects content that matches user preferences based on the terminal status history;
A morphological analysis unit that performs a morphological analysis on a character string included in the meta information of the content selected by the content selection unit;
The content recommendation system according to claim 2, further comprising: a morpheme correspondence determination unit that determines a morpheme selected from morphemes analyzed by the morpheme analysis unit as at least one of the associative word search words. The terminal status history is
Stores operation position history information indicating the history of positions operated on the screen,
The associative word search term determination unit
The content recommendation system according to claim 2, further comprising an operation position correspondence determination unit that determines at least one of the search words for the associative words based on the operation position history information. The terminal status history is
Stores operation time zone history information indicating the history of the time zone when the user operated the terminal device,
The associative word search term determination unit
The content recommendation system according to any one of claims 2 to 4, further comprising an operation time correspondence determination unit that determines at least one of the search words for associative words based on the operation time zone history information. An operation input unit for inputting an operation;
A terminal state information processing unit for generating terminal state information including the content of the operation input by the operation input unit as information indicating a predetermined state of the terminal device, and transmitting the generated terminal state information to the server;
A terminal device comprising: a recommended content list control unit that executes at least control for causing a display unit to display a recommended content list received from a server that communicates with the terminal device via a network. A terminal state history management unit that registers terminal state information transmitted from a terminal device that communicates with a server via a network in a terminal state history;
A server storage unit for storing the terminal state history;
Based on the terminal state history, recommended content to be recommended to the user of the terminal device is determined from recommended candidate content, and the recommended content list indicating the determined recommended content is transmitted to the terminal device A server comprising a department. A content recommendation method in a content recommendation system comprising a terminal device and a server that communicate with each other via a network,
The terminal device
An operation input step for inputting an operation;
A terminal state information processing step of generating terminal state information including the content of the operation input by the operation input step as information indicating a predetermined state of the terminal device, and transmitting the generated terminal state information to the server;
A recommended content list control step for performing at least control for displaying the recommended content list received from the server on the display unit,
The server
A terminal state history management step of registering the terminal state information transmitted from the terminal device in the terminal state history stored in the server storage unit;
Based on the terminal state history, recommended content to be recommended to the user of the terminal device is determined from recommended candidate content, and the recommended content list indicating the determined recommended content is transmitted to the terminal device A content recommendation method comprising steps. An operation input step for inputting an operation;
A terminal state information processing step of generating terminal state information including the content of the operation input by the operation input step as information indicating a predetermined state of the terminal device, and transmitting the generated terminal state information to the server;
A content recommendation method comprising: a recommended content list control step for performing at least control for causing a display unit to display a recommended content list received from a server that communicates with a terminal device via a network. A terminal state history management step of registering terminal state information transmitted from a terminal device communicating with a server via a network in a terminal state history stored in the server storage unit;
Based on the terminal state history, recommended content to be recommended to the user of the terminal device is determined from recommended candidate content, and the recommended content list indicating the determined recommended content is transmitted to the terminal device A content recommendation method comprising: On the computer,
An operation input step for inputting an operation;
A terminal state information processing step of generating terminal state information including the content of the operation input by the operation input step as information indicating a predetermined state of the terminal device, and transmitting the generated terminal state information to the server;
A program for executing a recommended content list control step for executing at least control for displaying a recommended content list received from a server communicating with a terminal device via a network on a display unit. On the computer,
A terminal state history management step of registering terminal state information transmitted from a terminal device communicating with a server via a network in a terminal state history stored in the server storage unit;
Based on the terminal state history, recommended content to be recommended to the user of the terminal device is determined from recommended candidate content, and the recommended content list indicating the determined recommended content is transmitted to the terminal device A program for executing steps and.

Images