JP6122270B2 - Information processing apparatus, control method thereof, and control program - Google Patents

Description

The present invention relates to an information processing system particularly suitable for a specific type of business. More specifically, the connection between people, such as SNS (social networking service) Te service smell to promote, about how to calculate the affinity between the event and the user.

  Conventionally, a community-type membership service called a social networking service (hereinafter referred to as “SNS”) that promotes and supports connections between people is known. SNS offers services such as a function to publish your profile and photos to members, a diary function that allows you to limit the scope of disclosure to specific friends, and a community function that allows you to decide themes such as hobbies and regions and interact on the bulletin board etc. Features that can be used in As one of such SNSs, a website disclosed in Non-Patent Document 1 is known.

  There are various service menus in the service. For example, as shown in FIG. 9, on the top page that is displayed when a member logs in to the service, the list 91 of activities that the member can perform within the service and the feed column 92 are attributed to the activities of other members. Then, a notice with a link distributed, that is, “feed” is displayed. When a member posts his / her daily events as “diary” content, the posting is delivered to other members as a linked feed. As contents that can be posted, there are various variations such as “calendar (plan)” and “photo” in addition to “diary”. The posting member can set the range in which the content posted by himself / herself is disclosed for each posting or for each content. As a setting method, for example, “open to all”, “open to friends”, “open to friends of friends” can be appropriately set. “Friends” are set by mutual approval between members. A plurality of levels can be set even among "friends".

  When various contents are posted, a feed to the effect that the posting has been made is distributed to other members who have a predetermined relationship with the posted member among the members to be published . The member who receives the feed can view the content posted via the link attached thereto. After browsing, it is possible to deepen communication between members by taking a reaction such as adding a comment to the post.

  Another menu is the “community” function. This allows members to post information on a bulletin board page in line with a specific theme, or to exchange information by viewing other members' posts. Posts to these “community” functions are also displayed on the member's top page and other frequently viewed pages, and the member can view the posted content by opening the link. You can also.

  Or, rather than semi-permanent pages like the “community” function, recruit and participate in specific projects (for example, a user ’s birthday party, tennis practice session, programming study session) An “event” function that facilitates mutual contact and information exchange between the users can also be provided as a menu. The “event” may be independent as one menu by itself, and may be recorded as one form of items to be input in the “calendar”, for example, and the name in the service does not matter. One user defines planning requirements such as the concept, date and place of planning, and the planner (founder) invites other users to this, or the invited user further invites other users. It is included if multiple users share their experiences online and / or offline.

When the initiator of an “event” or a user who plans to participate calls other users on the SNS, calls are often made to other users who are intimately related to themselves and who can easily recall. In some cases, information is not transmitted to an appropriate user due to the “event”. As a result, opportunities for new connections on networking services are lost unconsciously. However, an increase in invitation messages and feeds related to “events” that are not of interest to other users promotes annoyance to the service itself and leads to a decrease in user satisfaction. That is, if a function for selecting a user having a high affinity for the “event” based on a social graph of a user involved in the “event” is provided, such a problem is solved.

  Such technical proposals for optimizing information presented to service users or making appropriate recommendations have been made so far. For example, in Patent Document 1, as a method for optimizing information presented via a computer network, a first site information presentation space for presenting information on a first site that provides a plurality of information is displayed on the second site. Providing on the screen, selecting information from a plurality of information on the first site, presenting the selected information in the first site information presentation space on the second site, and first of a plurality of viewers on the second site An action history related to information on the site information presentation space is acquired, a recommendation rule specific to the second site is created using the action history, and the first site on the second site is created based on the recommendation rule specific to the second site. Selecting and updating the first site information presented in the information presentation space, and information on the first site information presentation space of a specific viewer of the second site; Action history is acquired, and a recommendation rule specific to a specific viewer is created using this action history, and is presented in the first site information presentation space on the second site based on the recommendation rule specific to the specific viewer A method for optimizing presentation information using a computer network, comprising: selecting and updating first site information. Alternatively, Patent Document 2 discloses a specific type of item based on a plurality of types of action logs of a user as a recommendation device that performs a recommendation reflecting various action logs that occur every day when making a recommendation about an item. A recommendation device that performs a recommendation on the item, a recommendation pattern storage unit that stores a recommendation pattern including an action log type and a recommendation technique suitable for use in the recommendation on the item for each user, and the recommendation pattern storage unit. A log extraction unit that acquires a recommendation pattern corresponding to a user to be recommended, and extracts an action log of an action log type in the recommendation pattern from a log storage unit that stores the plurality of types of action logs; and the log extraction unit Action log extracted by Based on, recommendation apparatus being characterized in that a recommendation processing means for performing recommendation processing relating to the item using the recommended approach in the recommendation pattern have been proposed.

Further, from the viewpoint of managing or adjusting the schedule of group members, as described in Patent Documents 3 to 4, a schedule management system in which an individual can input a schedule of members related to himself, or a specified user at a specified date and time A schedule management system that facilitates schedule adjustment by grasping the possibility of participation of the entire designated user based on the priority of the schedule information has been proposed.

JP2011-39909 JP2012-14659 JP-A-8-55158 Patent 4234408

  However, the optimization method disclosed in Patent Document 1 is excellent for optimization for the first time user, but optimization of presentation information that is optimized in a timely manner corresponding to event members and specifications that change from moment to moment. Is not done. In particular, the intimacy (affinity) between users also changes from moment to moment, and it has not yet been possible to present means for calculating the degree of affinity that users have for the event. In Patent Document 2, although the recommendation based on the action log that occurs every day is performed, the “field feature value” that dynamically reflects the basic parameters that characterize the event itself, the participating members, items related to the event, etc. A method for calculating and selecting a user with high affinity for the event based on the calculation result is not presented. Moreover, in patent document 3 thru | or patent document 4, although the management of the schedule by others or the structure which shows it with a priority is mentioned, it is from the viewpoint of the positioning of the event for the user who receives presentation. There is no mention of ranking.

An object of the present invention is to provide, in particular in the "Events" function provided in the networking service, and to dynamically determine the affinity between the event and the user.

In order to solve the above-described problem, the “information processing apparatus” according to one aspect of the present invention causes the affinity parameter of the event information related to the event to fluctuate due to an additional expression of participation in an event in which a plurality of users can participate. and recalculating means for the feature quantity of the field according to 該I vent whenever the recalculated in consideration of the affinity parameter after the change, the following variation in time the feature amount of the field is re-calculated by the recalculating means Specifying means for identifying a target user corresponding to user information whose degree of affinity matching has changed in relation to the affinity parameter of the field, and whenever the field feature quantity is recalculated by the recalculation means, Teisu Ru, respectively it calculate the affinity for the event for the user identified by the corresponding attribute value respectively compared to the specific means of the elements and the user information It comprises a constant means.
In order to solve the above-described problem, the “information processing apparatus control method” according to one aspect of the present invention provides an affinity of event information related to an event caused by an additional expression of participation in an event in which a plurality of users can participate. and recalculating steps of parameters every field recalculated in consideration of the affinity parameter after the change the feature amount of according to 該I vent varying each time the feature amount of play in the recalculation steps are recalculated A specifying step of identifying a target user corresponding to user information whose degree of affinity matching has changed in relation to the changed affinity parameter, and each time a field feature is recalculated in the recalculation step, the event of the target user to be identified in the identification step and corresponding to the attribute values versus respective feature amounts of elements and user information Including the calculated constant step Ru calculated Teisu, respectively it affinity for the bets.
In order to solve the above-described problem, the “control program” according to one aspect of the present invention causes the affinity parameter of the event information related to the event to fluctuate due to an additional expression of participation in the event that a plurality of users can participate in. a recalculation function to recalculate in consideration of the affinity parameter after the change characteristic of the field according to 該I vent every feature quantity of situ by the recalculation function after the change to the time it is re-calculated A function for specifying a target user corresponding to user information whose degree of affinity matching has changed in relation to the affinity parameter, and an element of the field feature value each time the field feature value is recalculated by the recalculation function and the affinity for the event for the user specified by the specifying function corresponding to the attribute values versus respective user information, respectively it calculated Teisu Calculated to achieve a constant function, to the computer of the information processing apparatus.

The following technical limitations may be added to the “information processing apparatus” according to one aspect of the present invention. Further, the same technical limitation may be added to the “information processing apparatus control method” and the “control program” according to one aspect of the present invention.
[1] A technical limitation may be added in which the specifying unit specifies the target user corresponding to user information whose degree of affinity matching has increased and has cleared a predetermined lower limit.
[2] In any one of the above inventions, the informing means for informing the user of the information outputs instruction information for informing the user of at least one of the users whose affinity is calculated by the calculating means. You may add the technical limitation further provided with an instruction | indication means.
[3] In any one of the above inventions, the presenting means for presenting at least one of the users whose affinity is calculated by the calculating means to the additional prospective participant who made the additional participation expression as a trigger The technical limitation of further comprising the above may be further added.

In the present invention,
“Event” means that one user defines planning requirements such as planning concept, date and time, place, etc., and the planner (founder) invites other users to this, or the invited user further multiple users by invite the user refers to one that is shared by either or both of the online / offline experience, does not matter referred to in the service. The “user” is typically a networking service member, but is not necessarily in the form of an SNS, and may be a member of a planning site that can participate in the short term.

The “affinity” may be defined as an index indicating whether one user has an attribute, an interest, or a social graph suitable for prompting participation in one event. In the present invention, the “affinity” is calculated based on the degree of matching between the “field feature value” of the event and the attribute value of the one user. The “field feature” is based on the “event properties” that are basic parameters such as the location, date, originator, and assumed participant profile of the event, and the group of users who are currently entering the event. It is calculated based on each element of an “affinity parameter” that is a variable to be measured and an “intelest parameter” that is a variable to be measured based on a variable element other than the participating user in the event.

The “affinity parameter” changes when a member newly joins. As a result, the “field feature” of the event also changes. “Intelest parameters” are typically given by each user who participated in an event “Let's give a gift to a specific user” (online digital content and offline content). Will be recorded cumulatively and will change accordingly. As a result, the “field feature” of the event also changes.

Furthermore, the nature of “events” varies widely. Specifically, when the event is a birthday party of a specific user, the closeness with the specific user should be emphasized. Further, in the case of a tennis practice event, a user who is expected to participate in the event is a user who is convenient in the scheduled date and time and place, so these basic parameters should be emphasized more. Alternatively, in the event of a lecture on a specific topic, whether or not there is an interest in the topic and the level of the topic are more important, not to mention the date and time and place. Therefore, as an embodiment of the present invention, there is provided a “weight correction unit” for giving a weight difference according to the category for the weight of the reference degree of each index of the event property, affinity parameter, and intelligent parameter described above. be able to. Specifically, it can be considered that an event balancer table that defines the weight of each index in calculating the affinity is referred to.

As an application example of the affinity service obtained based on the affinity calculation program as described above, an order of invited users to the event is generated and, for example, in the order of high affinity or a predetermined affinity according to the order. It is conceivable to generate recommendation information for presenting more users as invitation candidates. As a result, a user who is more appropriate for the event than the selection of the invited user based on the personal memory of the initiator or the user who has participated is presented, and the user can obtain event information more suitable for the user. This increases the possibility of participation of more appropriate users, leading to an increase in exchange opportunities as a networking service.

Alternatively, as another application example of the affinity service, it is possible to generate and distribute a feed related to the event to a user with a certain affinity or higher after assigning a user order based on the affinity. It is done. Thereby, the possibility that the information of the event is transmitted to the user suitable for the event increases, and the value of the service itself is improved for both the user and the event. Here, “feed distribution” refers to information such as “Notice of user A ’s birthday party”, which is displayed on the screen after the user logs in. As for the display mode, a difference may be provided in the display mode, for example, the size and color of characters, the thickness, the width of the feed area, and the like according to the order level of the user.

Furthermore, it is also possible to set an order in the feed itself separately from the order of the user and perform display control according to the order. In this case, each time a new feed is received, the order of the feeds can be recalculated to control the superiority or inferiority of the display mode between the feeds.

According to the present invention, since so that the affinity for additional the event each time the affinity parameter varies according to the event due to the participation assertion is calculated, the dynamic affinities between the event and the user It becomes possible to grasp. In addition, since the target user for calculating the affinity is limited to the user whose degree of affinity matching has changed, the amount of calculation can be reduced.
As a result, for example, a user who is more appropriate for the event is presented than the selection of the invited user based on the personal memory of the initiator or the joined user, and the user can obtain event information more suitable for him / her. As a result, the possibility of more appropriate users participating in the event increases, leading to an increase in exchange opportunities as a networking service.

It is a schematic diagram which shows the procedure which calculates the affinity (AR) of the event and a user in the affinity calculation program which concerns on this invention. It is a table | surface which shows an example of the weighting correction table in the affinity calculation program which concerns on this invention. It is a flow which shows the step which produces | generates the user recommendation based on affinity (AR). It is a flow which shows the step which performs feed delivery based on affinity (AR). It is a flow which shows the step of display order control when a new feed is delivered to one user in other examples. It is a schematic diagram which shows an example of the feed order (FL) table in one user. It is a schematic diagram which shows an example of the table which determines feed order (FL). It is a schematic diagram which shows an example of the correspondence of feed order (FL) and the display mode of a feed. It is a figure which shows an example of the screen shot of the top page of a SNS service. It is a network block diagram containing the server in which the affinity calculation program based on this invention integrated in the SNS service was installed.

Hereinafter, the present invention will be described in detail with reference to the drawings. In addition, the Example given below is only an example which embodied this invention to the last, and is not the meaning limited to this.

FIG. 1 is a schematic diagram showing a procedure for calculating an affinity (AR) between an event and a user in the affinity calculation program according to the present invention. The affinity calculation engine 10 functions when the affinity calculation program according to the present invention is installed in hardware and cooperates therewith. The installed hardware performs a computation process described later based on the function of communicating with other nodes such as an SNS server or a user terminal of a user participating in the SNS service via the network, and input data obtained by the communication. As long as it has a function of transmitting the output result to another node, a network server is typically assumed. FIG. 10 illustrates a network configuration diagram including a server in which the affinity calculation program according to the present invention incorporated in the SNS service is installed.

The SNS server 110 is connected to a plurality of user PCs 130 and a plurality of user portable terminals 160 (hereinafter sometimes referred to as “each user terminal”) via the network 120 to exchange information. The user portable terminal 160 is connected to the base station 150 wirelessly and is connected to the network 120 via a network device 140 such as a gateway. When receiving the SNS service, the user portable terminal 160 receives and installs the application in advance as necessary.

The SNS server 110 records each user's attribute information, activity information, and information on the relationship (social graph) between users. Based on the recorded information, the top screen of each user is generated and transmitted in response to a request from each user terminal. User logs in the service, that is, pages in the service viewed by the user, added comments and flags, generation of new pages, communication with other users, are recorded in the SNS server 110. Accessibility according to the relationship with the user is set in the recorded information, and whether or not other users can view is determined according to the accessibility and reflected on the screen display of the other user.

The user can know from the feed displayed on his / her top page that the activity of another user having a predetermined relationship has occurred. When one user requests display of his / her top page, information on the activities of other users related to the one user is retrieved, and a feed corresponding to the activity is generated. The generation opportunity of the feed is not limited to the request for browsing the top page of the user, but, for example, when an activity of the other user is generated, the feed is transmitted to another user having a predetermined relationship as viewed from the other user. It may be configured.

As described above, the feed based on the activity of each user is basically distributed depending on the social graph of the user. In other words, the extension of the reach area of the feed is limited by the social graph of the user (including the spread due to the chain of specific relationships viewed from the user). On the other hand, there are cases where it is not always reasonable to announce an event only for affinity within a service. From such a point of view, the affinity calculation server 170 that prescribes a field feature amount that is an eigenvalue of an event that can fluctuate and matches this with the information of each user can function effectively.

The affinity calculation server 170 is installed with the affinity calculation program according to the present invention. The affinity calculation server 170 receives the event article information posted by one user on the SNS server 110, receives the event article information from the SNS server 110, and calculates the field feature quantity (FQ). I do.

Here, the feature amount of the field will be described. The elements related to the event information 20 are basic attribute items (date and time of the event, venue, scheduled duration, etc.), the initiator (or responsible person) and the prospective participants, and the interests that characterize the nature of the event. , Can be disassembled for convenience. These factors are factors that influence each other from a different viewpoint when considering whether or not other users participate in the event. In the present invention, an element related to basic attribute matters is “event property (EP)”, an element related to the initiator (or responsible person) and a prospective participant is “affinity parameter (AP)”, and an element related to an interest item of the event is “ Each of them is quantified as an “intelest parameter (IP)”, and the total amount is grasped as a field feature amount 30 and used for calculating the degree of match with each user.

The numerical value here is not limited to quantification. For example, with regard to event properties, it means that each element of “location”, “date and time”, and “period of holding” is not one-dimensionalized but is grasped as a parameter to be monitored.

On the other hand, the user information 40 also includes profile information 41 relating to the user's attributes (address, age, sex, affiliation, etc.), affinity information 42 relating to the social graph formed by the user in the service, It can be grasped as each element of the interest information 43 relating to the matter of interest that can be grasped from the activity log or the like.

When the event information is posted, the field feature amount 30 calculated by the event property 21, the affinity parameter 22, and the intellectual parameter 23 is matched with the user information 40, and the affinity AR between the event and the user is determined. Is calculated. The calculation includes event / profile matching 11 that compares event property 21 and profile information 41, affinity matching 12 that compares affinity parameter (AP) 22 and affinity information 42, intelligent parameter (IP) 23, and intelligent information 43. This is performed by each of the functions of the intelligent matching 13 for comparing and the balancer 14 for correcting the weight of the result obtained by each matching. Calculation tables (not shown) for the event / profile matching 11, the affinity matching 12, the intelligent matching 13, and the balancer 14 are stored in the affinity calculation table DB 171 (see FIG. 10).

The event / profile matching 11 refers to the distance between the event location and the user's address, the user's schedule at the date / time of the event, and quantifies the user's ease of participation in the event. The affinity matching 12 quantifies the degree of coincidence between the event initiator or responsible person, the user who has expressed participation, and the social graph of the user being compared. If the user belonging to the social graph of the user is the founder or if many users are participating, the affinity matching 12 is highly evaluated. The intelligent matching 13 is the contents of an event, for example, if it is a study event, its contents and difficulty level, or if it is an artist's live event, the artist name and the activity log of the user to be compared The degree of coincidence with the text information related to the interests included is quantified.

Therefore, the field feature can be recalculated based on various triggers. For example, when a new user announces participation in an event, the feature amount of the field also varies as the affinity parameter (AP) varies. Or, if the event is a content that many users give a present to a single user, if it is a new opinion, or if it is an assorted gift, it will tend to be present as a whole by adding presents that are actually reserved and purchased Since the fluctuation occurs, the intellectual parameter (IP) fluctuates, and the field feature amount fluctuates. Alternatively, the event property (EP) is an invariable value in principle, but the event property (EP) may also change when the event holding point, for example, the venue is changed.

By the way, which of the above matching results is given more weight depends on the nature of the event. For example, when the event is a birthday party, the participation members are important, so the matching degree of the affinity matching 12 should be more important, whereas the venue / date of the tennis practice group is important because the event / profile is important The value of the matching 11 should be more emphasized. Or, since the content is important for an IT study group, it can be said that the degree of coincidence of the intellectual matching 13 should be more emphasized. From this point of view, the affinity calculation program 10 can be provided with a balancer 14 that performs weight correction for each matching.

FIG. 2 is a table showing an example of a weight correction table of the balancer 14 which is a weight correction unit for each element. As shown in the figure, each event has a different weight depending on its type. For example, in the “birthday party” event and “alumni association” event, it is highly likely that the affinity greatly affects each user as an element of whether or not to participate in the event, and is highly evaluated in calculating the affinity (AR). On the other hand, tennis practice sessions are often judged as to whether or not to participate depending on the venue and time. Therefore, the event parameter (EP) is set to be evaluated more heavily. The event type can be configured to select a category from existing options when the initiator inputs event information.

As described above, the affinity calculation server 170 compares the event information 20 and the user information 40 for each element, and calculates the affinity (AR). In the figure, each server (SNS server 110 and affinity calculation server 170) is described as a different node, but may be physically configured as the same machine device.

Next, a step of generating a user recommendation based on the affinity (AR) will be described with reference to FIG. The affinity calculation server 170 calculates the field feature amount 30 in response to the occurrence of the calculation opportunity (step S301). Next, the feature amount 30 of the field and the user information 40 of each user are compared for each element, and the affinity (AR) is calculated. Here, in order to define the extension of the user whose affinity (AR) is calculated, a lower limit value is set in each matching (event / profile matching 11, affinity matching 12, and intelligent matching 13) in each element. You may comprise so that affinity (AR) may be calculated only about the user who clears a lower limit in either of matching.

Thus, the affinity (AR) is calculated for each user (step S302). The affinity calculation server 170 generates a user order based on the affinity (step S303). Based on the generated user sequence, a user recommendation to be presented to the user according to the calculation trigger is generated. When the calculation trigger is posting of new event information, a user recommendation is generated as a candidate user list for inviting the event to the initiator. When the calculation trigger is a new user participation expression, a user recommendation is generated as a candidate user list to be further invited to the user (step S304).

Next, the step of performing feed generation based on the affinity (AR) will be described with reference to FIG. The affinity calculation server 170 calculates the field feature 30 in response to the occurrence of the calculation opportunity (step S401). Next, the feature amount 30 of the field and the user information 40 of each user are compared for each element, and the affinity (AR) is calculated. Here, in order to define the extension of the user whose affinity (AR) is calculated, a lower limit value is set in each matching (event / profile matching 11, affinity matching 12, and intelligent matching 13) in each element. You may comprise so that affinity (AR) may be calculated only about the user who clears a lower limit in either of matching.

Thus, the affinity (AR) is calculated for each user (step S402). Based on the calculated affinity, a feed display mode related to the event invitation is selected (step S403). Variations in the display mode of the feed include various ways such as fixing the display position higher, feeding with a photo, emphasizing a thick frame of the feed, feed display with bold characters, feed display with colored characters, and coloring of the feed background. In short, the higher the affinity, the easier it is to be recognized by the user. After such display mode selection is performed, the generated feed is distributed and displayed based on the selected display mode (step S404).

The calculated affinity (AR) is an index that directly indicates the relationship between the event information 20 and the user information 40. As described above, the feed related to the event information 20 depends on the affinity (AR). In the configuration in which the display control is performed, one of the elements that determine the feed order (FL), that is, the order of the feed itself that reflects the relationship with other elements that determine the display mode of each generated feed. It is positioned as one. In the following, an example of feed display order control that reflects the affinity (AR) will be described with reference to FIGS. 5 to 8 and FIG. In addition, about the structure which is common in the above-mentioned Example, the same code | symbol is used and the description is abbreviate | omitted.

FIG. 5 is a flow showing steps of feed display order control when a new feed is distributed to one user in another embodiment of the present invention. As in the above-described embodiment, when a trigger for affinity (AR) calculation such as posting of event information 20 or registration of a new participating user is generated, the field feature 30 relating to the event information 20 is calculated (step) S501). This is matched with the user information 40 of each user, the affinity (AR) is calculated, and the feed sequence (FL) initial value of the newly generated feed is calculated based on this affinity ( Step S502). Next, the affinity (AR) and feed order (FL) of the feed information stored for each delivery destination user of the feed are recalculated (step S503).

FIG. 6 is a table showing feed information stored in the feed DB 172 (see FIG. 10) for each user in the affinity calculation server 170 for each distribution destination user. FIG. 7 is a table showing a display order determination table stored in the display order determination table DB 173 in the affinity calculation server 170. FIG. 8 is a table showing an example of display correspondence according to the display order (FL).

The affinity calculation server 170 has display feed information for each user. As shown in FIG. 6, the feed ID of each feed, the event ID of the event that is the trigger of the feed, the affinity (AR) between the event and the user, and the feed generation time (t1) are one piece of information. The feed order (FL) is determined based on these information and the total number of feeds (n) scheduled to be displayed by the user.

FIG. 7 shows an example of a feed order (FL) determination table. As shown in this table, the smaller the total feed number (n) of the user, the higher the affinity (AR), and the difference between the feed generation time (t1) and the update time (t0). The shorter the elapsed time (t2) from the feed generation, the higher the feed order (FL) is determined. For example, for an affinity 60 feed that has passed 3 days from the initial generation, the feed order is “S” if the total number of feeds is 10 or less, but “B” if the total number of feeds exceeds 10 and is 20 or less. . In the case where the total number of feeds is 10 or less, the feed with affinity 60 is “S” in the feed order from the initial generation until the fourth day, but falls to “A” after the fifth day. Even if the affinity at the time of initial generation is low, if a large number of users belonging to the social graph of the user subsequently announce participation, the affinity parameter (AP) is improved, and as a result, the affinity (AR) is also improved, so the feed order is also improved. It will be rewritten higher.

By configuring the table in this way, the feed of the event with higher affinity for the user is displayed in a superior and conspicuous manner. On the other hand, the news value is increased due to the increase in the total number of feeds or the passage of time from the initial generation. Feeds that have decreased are now displayed at the bottom, and feeds for events under the specified conditions are hidden, which encourages the metabolism of the displayed feeds, and is always fresh and highly of interest to the user It is possible to realize a service in which a feed of events considered to be displayed according to the order.

Returning to FIG. 5, the recalculated affinity (AR) and feed order (FL) are overwritten and stored in the feed DB 172 (step S504), and feed display control information based on the new feed order (FL) is performed ( Step S505).

FIG. 8 shows an example of the correspondence of the display modes according to the feed order. By selecting a display method in a conspicuous manner that is more easily recognized by the user as the feed rank belongs to the higher rank, the opportunity for the user to participate in a more appropriate event is increased. Since it is possible to expand the participants in a form that does not depend only on the existing social graph, it is possible to create opportunities for interaction between users who have not interacted so far, and as a result, the service value for users is improved. .

Thus, the affinity calculation program according to the present invention, and the variations of the variables related to the event (e.g., add and the like changes Interesuto Participants to the event) affinity against to the event each time is calculated Therefore, it is possible to dynamically grasp the affinity between the event and the user . By applying this affinity to recommendation devices, feed distribution devices, and feed control devices that encourage further user participation, it is possible to know and participate in events that are more appropriate for the user and to have opportunities to interact with other users. It will increase and contribute to the improvement of traffic. As a result, the frequency of interaction between users increases, so that the utility value of services for individual users can be improved, and the commercial value as a medium for networking services can be further improved.

10 Affinity Calculation Program 11 Event / Profile Matching 12 Affinity Matching 13 Intelligent Matching 14 Balancer (Weighting Correction Unit)
20 Event information 30 Field feature 40 User information AR Affinity 110 SNS server 120 Network 130 User PC
160 User portable terminal 170 Affinity calculation server 171 Affinity calculation table DB
172 Feed DB
173 Display order determination table DB

Claims (6)

Due to the additional participants representations to multiple users can participate event considering the affinity parameter after the change characteristic of the field according to 該I vent whenever the affinity parameter of the event information according to the event variation Recalculation means to recalculate and
A specifying means for specifying a target user corresponding to user information whose degree of affinity matching has changed in relation to the changed affinity parameter each time a field feature is recalculated by the recalculating means;
Each time the field feature quantity is recalculated by the recalculation means, the element of the field feature quantity and the corresponding attribute value of the user information are respectively compared with respect to the event of the target user specified by the specification means . the affinity and San constant means it Ru respectively calculated Teisu,
An information processing apparatus comprising: Wherein the specifying means specifies the pre Symbol Target users the degree of pre-Symbol affinity matching corresponds to increased user information satisfies predetermined lower limit value,
The information processing apparatus according to claim 1 . Further comprising an instruction means to output the instruction information affinity causes the broadcast knowledge of the event for at least one of computed by Ruyu over The by said calculating means to informing means for informing information to the user,
The information processing apparatus according to claim 1 or 2 . Further comprising presenting means for presenting at least one of the user that the affinity by the calculating means the participation assertion triggered is calculated to additional prospective participant who the additional participants representations,
The information processing apparatus according to any one of claims 1 to 3 . Due to the additional participants representations to multiple users can participate event considering the affinity parameter after the change characteristic of the field according to 該I vent whenever the affinity parameter of the event information according to the event variation A recalculation step to recalculate and
A specifying step of identifying a target user corresponding to user information whose degree of affinity matching has changed in relation to the changed affinity parameter each time a field feature is recalculated in the recalculation step;
Each time the field feature amount is recalculated in the recalculation step, the element of the field feature amount and the corresponding attribute value of the user information are respectively compared with the event of the target user specified in the specification step. the affinity and San constant step it Ru respectively calculated Teisu,
A method for controlling an information processing apparatus. Due to the additional participants representations to multiple users can participate event considering the affinity parameter after the change characteristic of the field according to 該I vent whenever the affinity parameter of the event information according to the event variation And a recalculation function to recalculate
A specific function for identifying a target user corresponding to user information whose degree of affinity matching has changed in relation to the changed affinity parameter each time the field feature value is recalculated by the recalculation function;
Each time the field feature value is recalculated by the recalculation function, the element of the field feature value and the corresponding attribute value of the user information are respectively compared with the event of the target user specified by the specification function . the affinity and their respective Ru calculated Teisu calculated constant function,
Is a control program that causes a computer of an information processing apparatus to realize the above.

Images