Patent Application
20140195371
This application claims the benefit of Japanese Priority Patent Application JP 2013-001874 filed Jan. 9, 2013, the entire contents of which are incorporated herein by reference.
The present disclosure relates to an information processing apparatus, an information processing method, a program and a terminal apparatus.
In recent years, a variety of services through a network such as the Internet has been provided to users. For example, a social networking service (SNS) provides an occasion of communications among users through a network. A location-aware service provides diverse information that is associated with the current locations of users. Furthermore, many users utilize online stores to purchase products online.
Many online stores are provided with a scheme that recommends products to users. For example, if a user browses the detail information of a certain product, the information about products associated with the product is presented to the user as recommended products. Generally, a scheme for the recommendation is implemented by using some sort of recommendation algorithm, as typified by a collaborative filtering and content-based filtering that are described in Japanese Patent Laid-Open No. 2012-190061. The collaborative filtering is an algorithm based on the preference of a user, and determines a recommendation score using the information relevant to actions (for example, a purchasing, a viewing and listening, or a browsing) of other users who are similar in preference. The content-based filtering is an algorithm based on the attribute of an item such as a product, and determines a recommendation score based on the attribute of an item that is the object of an action of a user. Typically, items with high recommendation scores are selected as recommendation items to be presented to the user. Japanese Patent Laid-Open No. 2012-190061, in order to achieve an effective recommendation, proposes dynamically combining such two kinds of recommendation algorithms in response to a user's situation.
However, in the existing recommendation methods, there is not reflected the factor of word-of-mouth communication, which strongly affects the action of a user. Generally, word-of-mouth information is the information from other users who may or may not have any interest in sellers trying to sell items, and is one of important information for the user to determine his or her action such as a purchasing and a viewing and listening. However, it is troublesome for the user to actively collect word-of-mouth information. Furthermore, from a viewpoint of privacy protection, it is undesirable that the service side automatically collect pure word-of-mouth information, and distribute it among users.
Hence, it is desirable to implement a novel recommendation scheme that incorporates therein the factor of word-of-mouth communication and can resolve or reduce the above-described disadvantages.
According to an embodiment of the present disclosure, there is provided an information processing apparatus including: a recommendation unit configured to generate a recommendation information for a target user based on a preference information of at least one associated person having a social relationship through a communication service or a locational relationship with the target user; and a communication interface configured to provide the generated recommendation information to be sent to the target user.
According to another embodiment of the present disclosure, there is provided an information processing method including: generating a recommendation information for a target user based on a preference information of at least one associated person having a social relationship through a communication service or a locational relationship with the target user; and providing the generated recommendation information to be sent to the target user.
According to another embodiment of the present disclosure, there is provided a terminal apparatus forming part of a communication system, the communication system also including an information processing apparatus configured to provide recommendation information to the terminal apparatus, the terminal apparatus including: a circuitry configured to transmit and receive data signals via a network; send a request for recommendation information for a user of the terminal apparatus; and receive the recommendation information which is generated based on a preference information of at least one associated person having a social relationship though a communication service or a locational relationship with the user of the terminal apparatus.
According to another embodiment of the present disclosure, there is provided a method including: requesting a recommendation information for a target user from a server; and receiving the recommendation information from the server, wherein the recommendation information is generated based on a preference information of at least one associated person having a social relationship through a communication service or a locational relationship with the target user.
By the technology according to the present disclosure, it is possible to implement an effective recommendation scheme that incorporates therein the factor of word-of-mouth communication.
Hereinafter, embodiments of the present disclosure will be described in detail with reference to the appended drawings. Note that, in this specification and the appended drawings, structural elements that have substantially the same function and structure are denoted with the same reference numerals, and repeated explanation of these structural elements is omitted.
Descriptions will be given in the following order.
1. Outline of system
2. Configuration of server apparatus
2-1. Exemplary hardware configuration
2-2. Exemplary functional configuration
2-3. Exemplary process flow
3. Configuration of terminal apparatus
3-1. Exemplary hardware configuration
3-2. Exemplary functional configuration
3-3. Modifications
4. Exemplary recommendation scenario
4-1. First scenario
4-2. Second scenario
First, an outline of a recommendation system according to embodiments will be described using
The server apparatus 100 is an information processing apparatus that provides a recommendation function for recommending an appropriate item to a user. The server apparatus 100 is connected with the terminal apparatuses 200 through a network such as the Internet or a virtual private network (VPN). The item to be recommended by the server apparatus 100 may be any kind of items, such as a product to be sold at an online store, a video, picture or music content to be delivered through the network, advertising information, or a news article. The server apparatus 100 sends a recommendation result, in response to a recommendation request from an apparatus such as the terminal apparatus 200, or an application server, which is not shown in the figure.
The recommendation result to be generated by the server apparatus 100, typically, can contain a list of items whose recommendation scores, which are determined in a recommendation process, are high, or a list of the items and the recommendation scores. In embodiments, the server apparatus 100 determines a base score of the recommendation score (hereinafter, referred to as a basic recommendation score), in accordance with a known recommendation algorithm, which can include a collaborative filtering, a content-based filtering or a combination thereof. Then, the server apparatus 100 corrects the basic recommendation score using a correction recommendation score, and thereby generates an after-correction recommendation score. As described in detail later, the factor of word-of-mouth communication is incorporated in the correction recommendation score.
The terminal apparatus 200 is an information processing apparatus that is utilized by a user. The terminal apparatus 200 may be an information processing terminal such as a personal computer (PC), a smart phone, a personal digital assistant (PDA), a navigation apparatus or a game terminal, or may be a digital household electric appliance such as a television apparatus. The terminal apparatus 200 is not limited to the example of
In the example of
In this section, an exemplary configuration of the server apparatus 100 shown in
[2-1. Exemplary Hardware Configuration]
The communication I/F 101 is a communication interface that supports an arbitrary wireless communication protocol or wire communication protocol. The communication I/F 101 establishes a communication connection between the server apparatus 100 and the terminal apparatus 200. The input device 103 is a device through which an operator of the server apparatus 100 operates the server apparatus 100. The input device 103 can include a keyboard and a pointing device, for example. The display 105 includes a screen constituted by a liquid crystal display (LCD), an organic light-emitting diode (OLED) or a cathode ray tube (CRT), for example. The storage 107 is constituted by, for example, a high-capacity storage medium such as a hard disk, and stores various data that are within a database in the server apparatus 100. The memory 109 may be a semiconductor memory that can include a random access memory (RAM) and a read only memory (ROM), and stores programs and data for processing by the server apparatus 100. The bus 117 mutually connects the communication I/F 101, the input device 103, the display 105, the storage 107, the memory 109 and the processor 119. The processor 119 may be a central processing unit (CPU) or a digital signal processor (DSP), for example. The processor 119 executes the programs stored in the memory 109 or other storage media, and thereby activates various functions of the server apparatus 100, which will be described later.
[2-2. Exemplary Functional Configuration]
(1) Recommendation Unit
The recommendation unit 120 controls the execution of the recommendation process in the server apparatus 100. For example, once receiving a recommendation request from the terminal apparatus 200 through the communication I/F 101, the recommendation unit 120 starts an execution of the recommendation process. In the recommendation process, the recommendation unit 120 identifies the target user, for example, using a user ID contained in the recommendation request, and makes the associated-user selection unit 140 select associated users who have an association with the target user. Furthermore, the recommendation unit 120 makes the basic-score determination unit 152 and correction-score determination unit 154 determine a basic recommendation score SA and correction recommendation score SB for the target user, respectively. Next, the recommendation unit 120 makes the score correction unit 156 correct the basic recommendation score SA using the correction recommendation score SB and generate an after-correction recommendation score SC. Then, the recommendation unit 120 selects a recommended item based on the generated after-correction recommendation score SC, and sends, as a recommendation result, the information relevant to the recommended item to the terminal apparatus 200 through the communication I/F 101.
For example, the recommendation unit 120 may update the recommendation score at fixed intervals, and periodically send the new recommendation result to the terminal apparatus 200 of the target user. Alternatively, the recommendation unit 120 may update the recommendation score whenever a predetermined event is detected, and send the new recommendation result. Examples of the predetermined event can include a receipt of a recommendation update request, a change in the communication situation of the target user, a movement of the user, a new action of an associated user, or an increase or decrease in associated users.
(2) Recommendation Database
The recommendation DB 130 is a database in which various data to be used in the recommendation process are stored. In the example of
The user data 132 can contain a user ID, a nickname, attribute data (for example, age and sex), preference data (for example, categories of favorite items), position data, and communication situation data, for each of users who are registered in the recommendation system 10. The position data and communication situation data of users can be received from the individual terminal apparatuses 200, and stored in the recommendation DB 130. The item data 134 can contain an item ID, a name and attribute data (for example, categories), for each of the many items that are objects to be recommended. The data described here are just examples. That is, other types of data may be stored in the recommendation DB 130, and some of the above-described data may be omitted.
(3) Associated-User Selection Unit
The associated-user selection unit 140 selects one or more associated users who have an association with the target user, for determining the correction recommendation score. As a first criterion, the associated-user selection unit 140 may select users who are at the neighborhood of the target user as associated users. As a second criterion, the associated-user selection unit 140 may select associated users, based on the communication situation of the target user in a social network. As a third criterion, the associated-user selection unit 140 may select associated users, based on a recognition processing of a picture or voice acquired through an apparatus (for example, a camera or microphone mounted on the terminal apparatus 200) that the target user carries or wears. As a fourth criterion, the associated-user selection unit 140 may select, as associated users, users whom the target user designates through a user interface.
In the first criterion, the associated-user selection unit 140 may recognize users who are at the neighborhood of the target user, that is, associated users, based on position data that are collected from the terminal apparatuses 200 of the target user and other users. If the terminal apparatus 200 has a neighborhood-terminal detection function (for example, Wi-Fi Direct), the associated-user selection unit 140 may recognize, as associated users, users with neighborhood terminals that are detected by the terminal apparatus 200 of the target user. According to the first criterion, it is possible to incorporate in the recommendation score the factor of word-of-mouth communication from other users with whom the target user acts together in the real world, or other users who are in the place where the target user is visiting. Here, users (for example, family members of the target user) who are at the neighborhood of the target user for long periods may be excluded from associated users. Thereby, it is possible to avoid a loss of freshness of the recommendation result, caused by a continuous presentation of similar recommended items.
In the second criterion, the associated-user selection unit 140 may recognize, as associated users, users who are judged as having a high degree of intimacy with the target user, based on communication situation data that are collected from the terminal apparatus 200 of the target user. For example, users who frequently exchange messages with the target user can be judged as having a high degree of intimacy with the target user. Also, users who belong to the same community as the target user can be judged as having a high degree of intimacy with the target user. For example, the communication situation data can be generated from logs of an SNS or another service involving a social network in the terminal apparatus 200, and collected by the associated-user selection unit 140. According to the second criterion, it is possible to incorporate in the recommendation score the factor of word-of-mouth communication from other users in whom the target user is interested in the real world, or other users who are intimate with the target user. Here, the degree of intimacy between users may be regulated, by analyzing the contents of messages exchanged between the users using a natural language analysis technique. Thereby, it is possible to more accurately judge the degree of intimacy and select more appropriate associated users. The degree of intimacy is not limited to the above-described examples, and may be judged using a social graph that is acquired from an SNS.
In the third criterion, the associated-user selection unit 140 may recognize associated users, by applying a known personal recognition technique to a picture or voice that is acquired from the terminal apparatus 200 of the target user. In this case, the user data 132 stored in the recommendation DB 130 can contain face-picture data for individual users that are compared with the picture, or speech-feature data for individual users that are compared with the voice. According to the third criterion, it is possible to incorporate in the recommendation score the factor of word-of-mouth communication from other users with whom the target user acts or talks together in the real world, or other users in whom the target user is interested.
In the fourth criterion, the associated-user selection unit 140 may display a graphical user interface (GUI) for designating associated users on the screen of the terminal apparatus 200, and acquire the user IDs of one or more associated users through the displayed GUI. For example, associated users may be designated from a friend user list of the target user that is registered in an SNS. Alternatively, associated users may be designated from a list of associated user candidates that are extracted in accordance with the above-described first criterion, second criterion or third criterion. According to the fourth criterion, it is possible to select, as associated users, users from whom the target user wants to incorporate in the recommendation score the factor of word-of-mouth communication.
The above-described criteria for selecting associated users may be combined in any combination. Also, another selection criterion may be used. Furthermore, the associated-user selection unit 140 may provide, to the terminal apparatus 200, a GUI through which the target user designates a selection criterion in selection of associated users. For example, the associated-user selection unit 140 may display a list of the selection criteria on the screen of the terminal apparatus 200, and select associated users in accordance with the selection criterion designated by the target user. Thereby, it is possible to incorporate the factor of word-of-mouth communication, furthermore flexibly change the recommendation result in response to an intention of the user.
The associated-user selection unit 140 outputs a user ID list of associated users selected in this way, to the correction-score determination unit 154.
(4) Basic-Score Determination Unit
The basic-score determination unit 152 determines the basic recommendation score SA for the target user. The basic recommendation score SA can be determined by the basic-score determination unit 152, in accordance with a known recommendation algorithm, which can include the collaborative filtering, the content-based filtering or a combination thereof. For example, in the case of using the collaborative filtering, the basic-score determination unit 152 compares preference data contained in the user data 132 between the target user and other users, and adds a score to items that are the objects of the past actions of other users having a similar preference to the target user. The in-question other users can include also many users UG who are not associated users. In the case of using the content-based filtering, the basic-score determination unit 152 adds a score to items with a similar item attribute to an item that is the object of an action of the target user. Since details of the collaborative filtering and content-based filtering are known to those in the art, additional description is omitted here. The basic-score determination unit 152 may determine the basic recommendation score SA, in accordance with a recommendation algorithm different from the collaborative filtering and the content-based filtering. The basic-score determination unit 152 outputs the determined basic recommendation score SA to the score correction unit 156 and the recommendation unit 120.
(5) Correction-Score Determination Unit
The correction-score determination unit 154 determines the correction recommendation score SB, based on actions of the associated users selected by the associated-user selection unit 140. In embodiments, the correction-score determination unit 154 calculates the correction recommendation score SB, using the weight of each associated user and the rating value that is acquired for each associated user. Typically, the sum of the weights of all the associated users is 1.
For example, the correction-score determination unit 154 may determine the weight of each associated user, based on the position data of the target user.
In addition, for example, the correction-score determination unit 154 may determine the weight of each associated user, based on the communication situation of the target user.
Furthermore, the correction-score determination unit 154 determines the rating value of each item for each associated user. As an example, the rating value can be determined based on the action history of each of the associated users. For example, when a certain associated user views or listens to a video content or music content, the rating value of the content that was viewed or listened to increases. When a certain associated user browses or purchases a product at an online store, the rating value of the product increases. Actions of the associated users may be judged from operation logs of an application, such as an internet browser or a content player, in the terminal apparatuses 200, or may be judged from output data of a camera or sensor in the terminal apparatuses 200. The correction-score determination unit 154 may receive the action history generated in the terminal apparatus 200 and then determine the rating value based on the action history, or may receive the rating value determined in the terminal apparatus 200. In the following description, the action history or rating value received from the terminal apparatus 200 is referred to as the rating information. According to such methods, it is possible to automatically collect the rating value that can correspond to word-of-mouth information, without imposing a trouble of registering the word-of-mouth information on the associated users.
The correction-score determination unit 154 may attenuate the rating value determined based on the action history of each of the associated users, with time. In this case, the rating value of an item that was purchased, was viewed or listened to, or was browsed by an associated user, increases immediately after the action, and gradually decreases with time. The rating value may be attenuated with time in a linear manner or in a curved manner (for example, a Kaplan-Meier curve or a logistic curve). According to such methods, it is possible to adapt the correction recommendation score for a change in actions of the associated users, and successively update the recommendation result that reflects the factor of word-of-mouth communication.
As another example, the correction-score determination unit 154 may acquire the rating value that is explicitly designated by each of the associated users, as the rating information from the terminal apparatus 200. In this case, the correction-score determination unit 154 provides a GUI through which the associated users designate the rating value for each item, to the terminal apparatus 200 of the associated users. According to such a method, it is possible to reflect explicit evaluations of individual items by the associated users in the correction recommendation score.
The correction-score determination unit 154 can calculate the correction recommendation score SB for each item, by multiplying the weight and rating value determined in such a way for each item and then summing the products over all the associated users. Other than the above-described example, an equal weight may be used for all the associated users. Then, the correction-score determination unit 154 outputs the calculated correction recommendation score SB to the score correction unit 156.
The correction recommendation score SB may be a negative value. For example, the rating value of an item that an associated user dislikes may be determined as a negative value. Also, for example, the weight of an associated user who has a negative association with the target user (a person who is incompatible with the target user, or the like), may be determined as a negative value. The associated user who has a negative association may be explicitly designated by the target user, or may be judged by analyzing the contents of exchanged messages.
(6) Score Correction Unit
The score correction unit 156 generates the after-correction recommendation score SC, by correcting the basic recommendation score SA determined by the basic-score determination unit 152, using the correction recommendation score SB determined by the correction-score determination unit 154. In embodiments, the score correction unit 156 adds the product of the correction recommendation score SB and a synthesis ratio, to the basic recommendation score SA. The synthesis ratio is a ratio of the correction recommendation score SB to the basic recommendation score SA.
The first term of the left-hand side of the relational expression corresponds to the basic recommendation score SA. In the example of
The second term of the left-hand side of the relational expression corresponds to the product of the correction recommendation score SB and the synthesis ratio RB. As described above, the correction recommendation score SB is equal to the sum of the products, which result from multiplying the weight and rating value for each associated user, over all the associated users. In the example of
The right-hand side of the relational expression corresponds to the after-correction recommendation score SC. In the example of
In the example of
The score correction unit 156 may variably control the synthesis ratio RB. As an example, the score correction unit 156 may increase the synthesis ratio RB, while the target user participates in a specific community (for example, a community that is formed in an SNS). As another example, the score correction unit 156 may increase the synthesis ratio RB, while the target user is in a predefined specific place. Examples of the specific place include a place where many persons gather, such as a restaurant, a bar, a live hall, a stadium, a school, or a public hall. When a high value is set to the synthesis ratio RB, the proportion of the correction recommendation score SB included in the after-correction recommendation score SC increases, and the factor of word-of mouse communication has a greater effect on the recommendation result. Thereby, it is possible to increase a possibility that the experience about the same item is shared among users who participate in a community, or users who gather in the same place, and to animate the communication through a recommendation of an item.
[2-3. Exemplary Process Flow]
With reference to
Next, the correction-score determination unit 154 acquires the weight of each of the associated users that are selected by the associated-user selection unit 140 (step S20). Also, the correction-score determination unit 154 acquires the rating value of each item for each of the selected associated users (step S25). Then, the correction-score determination unit 154 calculates the correction recommendation score for each item, by summing the products of the acquired weights and rating values over all the associated users (step S30).
Next, the score correction unit 156 determines the synthesis ratio of the correction recommendation score to the basic recommendation score (step S35). Then, the score correction unit 156 corrects the basic recommendation score in accordance with the determined synthesis ratio, using the correction recommendation score calculated by the correction-score determination unit 154 (step S40).
Next, the recommendation unit 120 selects an item to be recommended based on the after-correction recommendation score generated by the score correction unit 156, and sends the recommendation result to the terminal apparatus 200 through the communication I/F 101 (step S45).
Thereafter, the recommendation unit 120 judges whether to end the recommendation process (step S50). For example, in the case where an application for displaying the recommendation result is shut down in the terminal apparatus 200, the recommendation unit 120 ends the recommendation process. If the recommendation process is continued, the flowchart returns to step S10. The recommendation result is updated periodically, or whenever a predetermined event is detected.
In this section, an exemplary configuration of the terminal apparatus 200 shown in
[3-1. Exemplary Hardware Configuration]
(1) Camera
The camera 201 includes an image-pickup element such as a charge coupled device (CCD) or a complementary metal oxide semiconductor (CMOS), and generates a pickup image. For example, in order to select the associated user, the camera 201 may pick up a user who is at the neighborhood of the target user. Furthermore, in order to recognize an action of the associated user, the camera 201 may pick up the associated user.
(2) Sensor
The sensor 203, typically, is a sensor module that can include a positioning sensor. For example, the positioning sensor may be a global positioning system (GPS) sensor that receives GPS signals to measure latitude, longitude and altitude, or may be a wireless-based sensor that measures a position based on wireless signals to be sent and received with a wireless access point. Position data generated by the sensor 203 can be collected by the server apparatus 100, for the selection of the associated user and the determination of the weight. The sensor 203 may include other types of sensors such as an electronic compass and an acceleration sensor.
(3) Input Device
The input device 205 is a device that a user uses for operating the terminal apparatus 200, or inputting information to the terminal apparatus 200. The input device 205 can include a touch sensor, a button, a switch, or a keypad, for example. The input device 205 may include a voice input module that detects a voice command given by a user as a user input. In the case where the terminal apparatus 200 is a wearable apparatus that includes an HMD, the input device 205 may include an eye-gaze detection module that detects an eye gaze of a user as a user input.
(4) Communication I/F
The communication I/F 207 is a communication interface that supports an arbitrary wireless communication protocol (for example, W-CDMA, WiMAX, LTE, LTE-A or wireless LAN) or wire communication protocol. The communication I/F 207 establishes a communication connection between the terminal apparatus 200 and the server apparatus 100. The communication I/F 207 may support a neighborhood-terminal detection function based on, for example, wireless LAN or Bluetooth (R).
(5) Memory
The memory 209 is constituted by a storage medium such as a semiconductor memory or a hard disk, and stores programs and data for processing by the terminal apparatus 200. Here, a part or a whole of programs and data to be described in this section may be acquired from an external data source (for example, a data server, a network storage, or an external memory), without being stored in the memory 209.
(6) Display
The display 211 includes a screen constituted by an LCD, an OLED or the like, and displays images. For example, the screen of the display 211 can display an application image for showing a recommendation result, and GUI images.
(7) Microphone
The microphone 213 is a voice input interface that collects voice given from a user or at the neighborhood of a user. For example, in order to select the associated user, the microphone 213 may collect voice of a user who is at the neighborhood of the target user.
(8) Bus
The bus 217 mutually connects the camera 201, the sensor 203, the input device 205, the communication I/F 207, the memory 209, the display 211, the microphone 213 and the processor 219.
(9) Processor
The processor 219 may be a CPU or a DSP, for example. The processor 219 executes the programs stored in the memory 209 or other storage media, and thereby activates various functions of the terminal apparatus 200, which will be described later.
[3-2. Exemplary Functional Configuration]
(1) Application Unit
The application unit 220 executes various applications that the terminal apparatus 200 has. The applications to be executed by the application unit 220 may be any kind of applications, such as an internet browser, a content player, an SNS client, an instant messenger, a VoIP client, a mailer, a television tuner, and an electronic book reader.
When an active application has a recommendation-result display function, the application unit 220 sends a recommendation request from the recommendation-result acquisition unit 234 to the server apparatus 100. Then, the application unit 220 displays the information about a recommended item on the screen, in accordance with a recommendation result that the recommendation-result acquisition unit 234 receives from the server apparatus 100.
(2) Situation Judgment Unit
The situation judgment unit 232 judges the communication situation and action of a user with the terminal apparatus 200. For example, in the case where the user with the terminal apparatus 200 is the target user, the situation judgment unit 232 may judge the communication situation of the target user and generate communication situation data in which the judged communication situation is described. The communication situation data can be generated from logs in an SNS or other services, for example. The communication situation data can contain, for example, login information to a social network for the target user, identification information in a community, and information relevant to a communication partner (for example, a user ID, communication time and communication frequency). The situation judgment unit 232 outputs the communication situation data generated in this way, to the recommendation-result acquisition unit 234. The situation judgment unit 232 may call the neighborhood-terminal detection function of the terminal apparatus 200, and outputs a user ID list of users with the detected neighborhood terminals, to the recommendation-result acquisition unit 234.
For example, in the case where the user with the terminal apparatus 200 is the associated user, the situation judgment unit 232 may judge an action of the associated user for an item, from operation logs in an application recorded in the application unit 220. The situation judgment unit 232 can judge, as the action of the associated user, a start and end of viewing and listening of a video content or music content, a browsing or purchasing of a product at an online store, or a browsing of a news article, for example. The situation judgment unit 232 may judge the action of the associated user, using a pickup image from the camera 201, sensor data from the sensor 203, or a voice inputted from the microphone 213, instead of operation logs in an application. The situation judgment unit 232 outputs a judgment result about such an action of the associated user, to the rating-information sending unit 236.
(3) Recommendation-Result Acquisition Unit
The recommendation-result acquisition unit 234 sends a recommendation request to the server apparatus 100, and receives a recommendation result from the server apparatus 100. The recommendation request can contain, in addition to the user ID of the target user, at least one of the position data of the target user, the communication situation data and a user ID list of associated user candidates. The associated user candidate may be a user with the neighborhood terminal, or may be a user who the target user designates through the GUI. Also, the recommendation request may contain the identifier of the selection criterion that is a criterion for selecting the associated user and that the target user can designate through the GUI. The recommendation-result acquisition unit 234, once receiving the recommendation result that the server apparatus 100 sends in response to the recommendation request, outputs the recommendation result to the application unit 220.
In the case where the target user has moved after the recommendation request was once sent to the server apparatus 100, the recommendation-result acquisition unit 234 may send the position data of the target user to the server apparatus 100, again. Also, in the case where the communication situation of the target user has changed, the recommendation-result acquisition unit 234 may send the communication situation data of the target user to the server apparatus 100, again. In addition, in the case where the associated user candidate has changed, the recommendation-result acquisition unit 234 may send the user ID list of associated user candidates to the server apparatus 100, again. Such data sending may be performed periodically.
(4) Rating-Information Sending Unit
The rating-information sending unit 236 sends the above-described rating information to the server apparatus 100. The rating-information sending unit 236 may generate the rating information, based on the judgment result for the action of the associated user, which is inputted from the situation judgment unit 232. The rating information can contain the action history of the associated user, or the rating value determined based on the action history. Alternatively, the rating-information sending unit 236 may generate the rating information containing the rating value that the associated user designates through the GUI.
(5) Recommendation Score Switching
As an example, the recommendation-result acquisition unit 234 may provide, to the target user, a user interface for switching the recommendation score that is the basis of the recommendation result, between the basic recommendation score and the after-correction recommendation score.
With reference to the right side of
By switching the recommendation score through such a user interface as shown in
[3-3. Modifications]
Some of the functions of the server apparatus 100 described using
In this section, there will be described exemplary recommendation scenarios that are implemented using the above-described server apparatus 100 and terminal apparatus 200.
[4-1. First Scenario]
With reference to
Next, the user UA starts an application having a recommendation-result display function, on the terminal apparatus 200 (step S120). For example, the terminal apparatus 200 of the user UA calls a neighborhood-terminal detection function, and detects neighborhood users who are at its own neighborhood (step S122). Then, the terminal apparatus 200 of the user UA sends a recommendation request to the server apparatus 100 (step S124). The recommendation request to be sent at this time can contain, for example, the position data of the user UA and a list of the user IDs of the neighborhood users (associated user candidates), in addition to the user ID of the user UA who is the target user.
Next, the server apparatus 100, once receiving the recommendation request, executes the recommendation process described using
Next, the terminal apparatus 200 of the user UA, once receiving the recommendation result from the server apparatus 100, displays the information about the recommended items on the screen, in accordance with the received recommendation result (step S130). In the example of
Next, with reference to
Next, the user UF2 comes close to the user UA, and starts a playback of an item IT22 in the terminal apparatus 200 of the user UF2 (step S150). In response to the playback start of the item IT22, the terminal apparatus 200 of the user UF2 generates rating information, and sends the generated rating information to the server apparatus 100 (step S152). The server apparatus 100 acquires a rating value for the user UF2 from the received rating information (step S154).
The terminal apparatus 200 of the user UA, for example, periodically executes the neighborhood-terminal detection function, and detects the user UF2 who is a neighborhood user at its own neighborhood (step S162). Then, the terminal apparatus 200 of the user UA sends a recommendation update request to the server apparatus 100 (step S164). The recommendation update request to be sent at this time can contain, for example, the latest position data of the user UA and a list of the user IDs of the neighborhood users in which the user IDs of the users UF1 and UF2 are described.
Next, the server apparatus 100, once receiving the recommendation update request, executes the recommendation process, again (step S166). Unlike the recommendation process in step S126, two persons, the users UF1 and UF2 are associated users in step S166. The rating value for the user UF1 has been already acquired in step S144. The rating value for the user UF2 has been already acquired in step S154. The rating value for the user UF1 that has been acquired at an earlier time, may be attenuated with time. The server apparatus 100 corrects the basic recommendation score using the correction recommendation score that is determined from these rating values, and selects recommended items based on the after-correction recommendation score. Then, the server apparatus 100 sends a recommendation result to the terminal apparatus 200 of the user UA (step S168).
Next, the terminal apparatus 200 of the user UA, once receiving the recommendation result from the server apparatus 100, updates the information about the recommended items on the screen, in accordance with the received recommendation result (step S170). In the example of
[4-2. Second Scenario]
With reference to
Next, the user UA starts an application having a recommendation-result display function, on the terminal apparatus 200 (step S220). For example, the terminal apparatus 200 of the user UA generates communication situation data, and sends the generated communication situation data to the server apparatus 100 (step S222). Also, the terminal apparatus 200 of the user UA sends a recommendation request to the server apparatus 100 (step S224). Here, the communication situation data may be contained in the recommendation request, instead of being sent separately from the recommendation request.
Next, the server apparatus 100, once receiving the recommendation request, sends a rating request to each of the terminal apparatuses 200 of the user UF1 and user UF2 who are selected as associated users (step S226). The terminal apparatus 200 of the user UF1 sends rating information to the server apparatus 100, in response to the rating request (step S228). Similarly, the terminal apparatus 200 of the user UF2 sends rating information to the server apparatus 100, in response to the rating request (step S230). Then, the server apparatus 100 executes the recommendation process described using
Next, the terminal apparatus 200 of the user UA, once receiving the recommendation result from the server apparatus 100, displays the information about the recommended items on the screen, in accordance with the received recommendation result (step S236). In the example of
Next, with reference to
Next, the server apparatus 100, once receiving the recommendation update request, executes the recommendation process, again (step S246). Unlike the recommendation process in step S232, since the user UA does not participate in the community in step S246, the server apparatus 100 can use a lower synthesis ratio. Then, the server apparatus 100 sends a recommendation result to the terminal apparatus 200 of the user UA (step S248).
Next, the terminal apparatus 200 of the user UA, once receiving the recommendation result from the server apparatus 100, updates the information about the recommended items on the screen, in accordance with the received recommendation result (step S250). In the example of
Additionally, the present technology may also be configured as below.
(1) An information processing apparatus including:
a recommendation unit configured to generate a recommendation information for a target user based on a preference information of at least one associated person having a social relationship through a communication service or a locational relationship with the target user; and
a communication interface configured to provide the generated recommendation information to be sent to the target user.
(2) The information processing apparatus of (1),
wherein the recommendation information includes at least one of a video, picture, music content, advertising information, and a news article.
(3) The information processing apparatus of (1) or (2), further including:
an associated person selection unit configured to select the at least one associated person.
(4) The information processing apparatus of any of (1) through (3), wherein the associated person selection unit selects the at least one associated person based on a user ID of the target user.
(5) The information processing apparatus of any of (1) through (4), wherein the associated person selection unit selects the at least one associated person such that each one of the at least one associated person is physically located within a predetermined vicinity of the target user.
(6) The information processing apparatus of any of (1) through (5), wherein the associated person selection unit selects the at least one associated person based on a communication status of the target user through the communication service.
(7) The information processing apparatus of any of (1) through (6), wherein the communication status is a frequency of communication between the target user and another user within the communication service.
(8) The information processing apparatus of any of (1) through (7), wherein each one of the at least one associated person is a registered friend of the target user within a social media service.
(9) The information processing apparatus of any of (1) through (8), wherein the recommendation unit generates the recommendation information based on the preference information of the at least one associated person and a preference information of the target user.
(10) The information processing apparatus of any of (1) through (9), wherein the recommendation unit generates the recommendation information based on a detection of a triggering event, the triggering event being at least one of a receipt of a request to update the recommendation information, a detected change in a communication status of the target user, a detected movement of the target user, a detected action made by one of the at least one associated person, and a detected change in a number of the at least one associated person.
(11) The information processing apparatus of any of (1) through (10), wherein the at least one associated person has a social relationship with the target user, and each one of the at least one associated person has previously communicated with the target user within a social media service.
(12) The information processing apparatus of any of (1) through (11), wherein the recommendation unit generates the recommendation information by determining a basic recommendation score for the target user, determining a correction recommendation score based on the preference information of the at least one associated person, correcting the basic recommendation score by using the correction recommendation score, and generating the recommendation information based on the corrected basic recommendation score.
(13) The information processing apparatus of any of (1) through (12), wherein the correction recommendation score is determined by weighting each of the at least one associated person.
(14) The information processing apparatus of any of (1) through (13), wherein the correction recommendation score changes based on a situation of the target user.
(15) An information processing method including:
generating a recommendation information for a target user based on a preference information of at least one associated person having a social relationship through a communication service or a locational relationship with the target user; and
providing the generated recommendation information to be sent to the target user.
(16) A non-transitory computer-readable medium having embodied thereon a program, which when executed by a computer causes the computer to execute a method, the method including:
generating a recommendation information for a target user based on a preference information of at least one associated person having a social relationship through a communication service or a locational relationship with the target user; and
providing the generated recommendation information to be sent to the target user.
(17) A terminal apparatus forming part of a communication system, the communication system also including an information processing apparatus configured to provide recommendation information to the terminal apparatus, the terminal apparatus including:
a circuitry configured to
requesting a recommendation information for a target user from a server; and
receiving the recommendation information from the server,
wherein the recommendation information is generated based on a preference information of at least one associated person having a social relationship through a communication service or a locational relationship with the target user.
(21) A non-transitory computer-readable medium having embodied thereon a program, which when executed by a computer causes the computer to execute a method, the method including:
requesting a recommendation information for a target user from a server; and
receiving the recommendation information from a server,
wherein the recommendation information is generated based on a preference information of at least one associated person having a social relationship through a communication service or a locational relationship with the target user.
(22) An information processing apparatus comprising
a score correction unit to correct a basic recommendation score using a correction recommendation score, the basic recommendation score being determined for a user by a recommendation algorithm, the correction recommendation score being determined based on actions of one or more persons who have an association with the user.
(23) The information processing apparatus of (22), further comprising
a selection unit to select the one or more persons for determining the correction recommendation score.
(24) The information processing apparatus of (22) or (23), wherein the selection unit selects the one or more persons, based on a communication situation of the user in a social network.
(25) The information processing apparatus of any of (22) through (24), wherein the selection unit selects the one or more persons who are at a neighborhood of the user.
(26) The information processing apparatus of any of (22) through (25), wherein the selection unit selects the one or more persons, based on a recognition processing of a picture or a voice acquired through an apparatus that the user carries or wears.
(27) The information processing apparatus of any of (22) through (26), wherein the selection unit selects the one or more persons whom the user designates through a user interface.
(28) The information processing apparatus of any of (22) through (27), wherein the selection unit provides a user interface through which the user designates a selection criterion in selection of the one or more persons.
(29) The information processing apparatus of any of (22) through (28), further comprising
a correction-score determination unit to determine the correction recommendation score based on actions of the one or more persons.
(30) The information processing apparatus of any of (22) through (29), wherein the correction-score determination unit calculates the correction recommendation score, using a weight of each person and a rating value acquired for each person.
(31) The information processing apparatus of any of (22) through (30), wherein the weight is determined based on a communication situation or position data of the user, or is designated by the user.
(32) The information processing apparatus of any of (22) through (31), wherein the rating value is determined based on an action history of each of the one or more persons, or is designated by each of the one or more persons.
(33) The information processing apparatus of any of (22) through (32), wherein the rating value is attenuated with time, after the rating value is determined based on an action history of each of the one or more persons.
(34) The information processing apparatus of any of (22) through (33), wherein the score correction unit variably controls a ratio of the correction recommendation score to the basic recommendation score.
(35) The information processing apparatus of any of (22) through (34), wherein the score correction unit increases the ratio of the correction recommendation score to the basic recommendation score, while the user participates in a specific community.
(36) The information processing apparatus of any of (22) through (35), wherein the score correction unit increases the ratio of the correction recommendation score to the basic recommendation score, while the user is in a specific place.
(37) The information processing apparatus of any of (22) through (36), wherein the recommendation algorithm includes at least one of an algorithm based on a preference of a user and an algorithm based on an attribute of an item.
(38) An information processing method to be executed by an information processing apparatus, the method comprising
correcting a basic recommendation score using a correction recommendation score, the basic recommendation score being determined for a user by a recommendation algorithm, the correction recommendation score being determined based on actions of one or more persons who have an association with the user.
(39) A program for causing a computer that controls an information processing apparatus to function as:
a score correction unit to correct a basic recommendation score using a correction recommendation score, the basic recommendation score being determined for a user by a recommendation algorithm, the correction recommendation score being determined based on actions of one or more persons who have an association with the user.
(40) A terminal apparatus comprising:
a communication interface to communicate with a server apparatus that corrects a basic recommendation score using a correction recommendation score, the basic recommendation score being determined for a user by a recommendation algorithm, the correction recommendation score being determined based on actions of one or more persons who have an association with the user; and
a control unit to display information of a recommended item on a screen, in accordance with a recommendation result that is received from the server apparatus through the communication interface,
wherein the control unit sends a list of the one or more persons to the server apparatus, and receives a recommendation result from the server apparatus, the recommendation result being based on the correction recommendation score that is determined using the sent list.
(41) The terminal apparatus of (40),
wherein the correction recommendation score is calculated using a weight of each person and a rating value acquired for each person, and
wherein the control unit informs the sever apparatus of the weight through the communication interface, the weight being determined based on a communication situation or position data of the user, or designated by the user.
(42) A terminal apparatus incorporating a circuitry configured to transmit and receive data signals, the terminal apparatus forming part of a communication system, the communication system also including:
an information processing apparatus comprising:
a recommendation unit configured to generate a recommendation information for a target user based on a preference information of at least one associated person having a social relationship or a locational relationship with the target user; and
a communication interface configured to provide the generated recommendation information to the target user;
wherein when the information processing apparatus receives, from the terminal apparatus, a request for the recommendation information for the target user, the recommendation unit generates the recommendation information and the generated recommendation information is provided to the terminal apparatus through the communication interface.
(43) The terminal apparatus of (42), wherein the information processing apparatus further comprises an associated person selection unit configured to select the at least one associated person.
(44) The terminal apparatus of (42) or (43) wherein the associated person selection unit selects the at least one associated person based on a user ID of the target user.
(45) The terminal apparatus of any of (42) through (44), wherein the associated person selection unit selects the at least one associated person such that each one of the at least one associated person is physically located within a predetermined vicinity of the target user.
(46) The terminal apparatus of any of (42) through (45), wherein the associated person selection unit selects the at least one associated person based on a communication status of the target user.
(47) The terminal apparatus of any of (42) through (46), wherein the communication status is a frequency of communication between the target user and respective ones of the at least one associated person.
(48) The terminal apparatus of any of (42) through (47), wherein each one of the at least one associated person is a registered friend of the target user within a social media service.
(49) The terminal apparatus of any of (42) through (48), wherein the recommendation unit generates the recommendation information based on the preference information of the at least one associated person and a preference information of the target user.
(50) The terminal apparatus of any of (42) through (49), wherein the recommendation unit generates the recommendation information based on a detection of a triggering event, the triggering event being at least one of a receipt of a request to update the recommendation information, a detected change in a communication status of the target user, a detected movement of the target user, a detected action made by one of the at least one associated person, and a detected change in a number of the at least one associated person.
(51) The terminal apparatus of any of (42) through (50), wherein the at least one associated person has a social relationship with the target user, and each one of the at least one associated person has previously communicated with the target user within a social media service.
(52) The terminal apparatus of any of (42) through (51), wherein the recommendation unit generates the recommendation information by determining a basic recommendation score for the target user, determining a correction recommendation score based on the preference information of the at least one associated person, correcting the basic recommendation score by using the correction recommendation score, and generating the recommendation information based on the corrected basic recommendation score.
| Number | Date | Country | Kind |
|---|---|---|---|
| 2013-001874 | Jan 2013 | JP | national |
