KR101849658B1 - Targeting stories based on influencer scores - Google Patents

Abstract

The news describing the activities performed by the interacting user is distributed to the user being browsed according to the influence score for the user being viewed. Each influencer score can be calculated based at least in part on the influencing of the user on the user being browsed and on the influences of the user being browsed and on the users who are connected to the user being browsed. Based on the determined influencer score, at least one of the viewing users may be provided with a description of the activity performed by the interacting user.

Description

{TARGETING STORIES BASED ON INFLUENCER SCORES}

FIELD OF THE INVENTION The present invention relates generally to social networking, and more particularly to computing news influencer scores for users of social networking systems and targeting news describing user activity based on influencer scores.

Social utilities that keep track of and enable connections between social networks or users (including people, businesses and other entities) are becoming popular in recent years. In particular, social networking systems allow users to communicate information more efficiently. For example, a user may post contact information, background information, job information, hobbies, and / or other user-specific data about a location associated with a user on the social networking system. Other users can then review the posted data by browsing the user profile and searching the profile containing the specific data. The social networking system also allows users to associate them with other users and thereby create a connected web between users of the social networking system. This connection between users can be used by the social networking system to provide more relevant information for each user in terms of the interests that the user has expressed.

Social networking systems typically include a system for connecting users with content likely to be associated with each user. For example, a user may be grouped according to one or more common attributes in their profile, such as geographic location, employer, type of occupation, age, music symbol, interest or other attributes. The user or outside party of the social networking system may then customize or target the information delivery using such groups so that information that a particular group may have particular interest may be communicated to that group.

Advertisers have been trying to increase the impact of this information on their members by targeting ads to members who have the best interest in advertising. For example, a social networking system may display a banner ad for a concert to members living near a concert venue where the band includes a sense of intimacy with the performance band in the social networking system profile. However, this attempt is no different from targeting advertisements that exist in many different situations. Advertisers have not yet utilized the relationships and connections between members of social networking systems in a meaningful way that increases user engagement with advertising, advertising products or brands.

Embodiments of the invention relate to determining the influencer scores of a plurality of users of a social networking system. In one embodiment, the influencer score is based at least in part on the influence of the first user on one or more users associated with the first user and on the impact scores of one or more users. For example, the influencer score of user John may be based on John's influence on another user Bob and the influence score of user Bob. Bob's influence scores can then be based on Bob's influence on users Joanne and Roger and Joanne and Roger's influence scores. In one embodiment, the influence of the first user on a user associated with the first user may be based on the number and type of interactions performed by the connected user in the content of the first user. For example, a first user may share a link to a web page via a social networking system. The user associated with the first user can then create a comment on the sharing of that link. The comment of the connected user may be used to determine the influence of the first user on the connected user.

In one embodiment, the news describing the activities performed by the interacting user is distributed to one or more browsing users according to a set of influencer scores of the browsing users. In this embodiment, each influencer score may be associated with a category assigned to an activity performed by the interacting user. For example, an interactive user may share a link to an article posted on an investment web site. Since the link is associated with an article from an investment website, the activity of the interacting user can be assigned to a financial category. Thus, the influence score of the user being browsed may be based on the influencing user's influence on each of his or her connections to the financial category and the influencer score of each connection on the same financial category. Based on the influential score of the user being browsed and also in some instances the influence of the interacting user on the user being browsed, the news describing the activity performed by the interacting user may be provided to the user being browsed .

By sending the news to the viewing user on a partial basis of the influencer score, the embodiments can improve overall effectiveness of the news. More specifically, because the user being browsed is provided with news based on the influence of the user being browsed for other users for a particular category, subsequent interaction (e.g., sharing of news) with the news by the viewing user, May also be interested in the news and enable effective targeting of the news to other users who are likely to interact with the news.

The features and advantages described in the foregoing detailed description of the invention are not all inclusive. Many additional features and advantages will be apparent to those skilled in the art in view of the drawings, detailed description, and claims of the invention.

1 is a diagram of a process for providing news based on an influencer score in accordance with one embodiment of the present invention.
2A is an upper layer block diagram illustrating a system environment suitable for operation of a social networking system according to an embodiment of the present invention.
2B is a block diagram of various components of a social networking system in accordance with an embodiment of the present invention.
3 is a flow diagram of a process for providing news to a viewing user based on an influencer score in accordance with one embodiment of the present invention.
The drawings illustrate various embodiments of the present invention by way of example only. Those skilled in the art will readily appreciate that alternative embodiments of the configurations and methods described herein may be utilized without departing from the principles of the invention disclosed herein through the following description.

summary

The social networking system provides the user with the ability to communicate and interact with other users of the social networking system. In use, a user joins a social networking system and then adds a connection with a number of other users that the user wishes to connect to. The terms "friend," " follower, "and" fan, " as used herein, refer to any other user who has established a connection, association or relationship through a social networking system. Connections may be explicitly added by a user, for example, the user may select a particular other user to be a friend, or may be based on the common characteristics of the users (e.g., users who are graduates of the same educational institution) It can be automatically generated by a social networking site. In social networking systems, connections are usually in both directions, but not necessarily so, the terms "user" and "friend" depend on criteria of reference. For example, if Bob and Joe are both users and connected in a social networking system, Bob and Joe are friends. The connection between users can be a direct connection; However, some embodiments of the social networking system allow the connection to be indirect through one or more connection levels. Also, the term friend does not necessarily mean that users are actually friends in real life (i.e., in general, one of the users is a business or another entity); A friend simply refers to a connection in a social networking system. As used herein, the term connection may also be used to generally refer to a user's friend, follower, or fan.

In addition to interacting with other users, social networking systems provide users with the ability to interact with or perform activities on various types of social networking objects supported by the social networking system. A social networking object may include a profile, an application (e.g., a game executable within a social networking system), an event (e.g., a page that represents a concert where the user can attend), a group (E.g., pages that constitute a particular entity's presence in a social networking system), locations associated with the user (e.g., "Palo Alto, California, USA" (E.g., pages containing content), user-generated content items (e.g., user posts), physical or digital item representations, concepts, and the like. A user can interact with a social networking object by associating with or interacting with the object. For example, a user can interact with an object by joining a group, participating in an event, checking into a location, becoming a fan of an organization's fan page, "liking" a fan page, . These are just some examples of objects that a user can act on in a social networking system, and many others are possible. In addition, the user interaction may include items of user generated content. For example, a user may interact with the fan page by posting to the company ' s fan page. A post may include user generated comments that provide a user's view of the company's product.

In one embodiment, the influencer score is determined for one or more browsing users, and the news describing the activity performed by the interacting user is provided to the browsing users based on the influencer score. As used herein, an influencer score of a user being viewed may indicate a value indicative of the influence of the user being browsed on his or her connection relationship and his or her influence score. For example, an influencer score can take into account how influential the user being viewed is on one of the friends and also the score of the friend's influence on her own friends.

By providing news to the viewing user based on the influencer score of the user being viewed, embodiments can improve overall effectiveness of the news. More specifically, since the user being browsed is provided with news based on the influence of the user being browsed to other users, future interaction (e.g., sharing of news) with the news by the user being browsed may also It may enable effective targeting of the news to other users who are interested and likely to interact with the news.

In one embodiment, the influencer score set of the users being viewed may be associated with a particular category assigned to an activity performed by the interacting user. For example, an interacting user may share a link to a comment posted on a baseball-themed website. Since the link is related to the above-described website, the activity of the user in interaction can be assigned to the sports category. Also, each determined influencer score of the users being viewed may be associated with a sports category. As a result, the influencer score can more accurately reflect the impact that the interacting user's ability to perform may have on other users.

In one embodiment, each influencer score of the user being browsed may be a function of the influencing value of the user being browsed on his or her connection relationships and the influential score of the connection relationships. In this embodiment, the influencer score of the user being viewed may be calculated using the following equation:

In the above formula, F (v) represents the influential score of the predetermined viewing user v. G (v, c) represents the influence value of the user (v) on the connection relationship (c), where the user (v) being browsed is associated with i connection relationships. F (c) represents the influence score of the connection relationship (c).

In general, the influence value G (v, c) indicates the amount of influence that the viewing user has on the connection relationship. In one embodiment, the influence value is the number of interactions performed by the connection relationship in the content of the user being browsed (e.g., content generated by the user being browsed, news describing the activity performed by the browsing user, etc.) And type. In one aspect, if the number of interactions performed by the connection relationship in the content of the user being browsed is relatively high, the influence value may also have a relatively high value. In yet another aspect, the influence value can be determined according to the weight of the type of interaction performed by the connection relationship, other types of interactions being weighted differently. For example, if a connection does not like the activity performed by the user being viewed, but the user comments on the activity, the impact value may be higher. In one embodiment, the impact value may also be calculated based on the type of relationship between the user being viewed and the connection relationship. For example, the influence value may be higher if the connection is a friend of the user being browsed, rather than a fan of the user being browsed. In one aspect, the content of the browsing user that is considered when calculating the impact value may be limited to content related to the same category as the activity performed by the interacting user.

The influential score F (c) of a connection relation of a user who is being browsed is calculated by the following equation: F (c) = F (k) ) Can be used as a basis. The influence value of the connection relationship can be determined in a manner similar to the determination of the influence value G (v, c) of the user being browsed.

In one embodiment, the influencer score of the user being viewed may be computed as the sum of the set of influential user scorer components of the user being browsed, and each influencer score component is weighted against the corresponding influencer score F (c) And the influence value G (v, c). In one embodiment, the at least one browsing user may be provided with a post describing the activity performed by the interacting user, the posting being associated with at least one viewing user, ≪ / RTI > The influential score component of the interacting user may be based on the calculated influential score F (v) of the user being browsed and the value of the interacting user's influence on the user being browsed. The impact value of the interacting user can generally indicate the amount of influence the interacting user has on at least one viewing user.

In one embodiment, if the interactive user's score component of the interacting user for the user being viewed exceeds a predetermined threshold, then at least one viewing user may be provided with the news. In another embodiment, if the influencer score component corresponding to the user being viewed is larger than other influencer score components corresponding to the user being viewed, at least one viewing user may be provided with the news. For example, the interacting friend 1 of the user being viewed may have liked the article posted on the sports web page. Friend 2 of the user being interacted with may be sharing an article posted on a cooking blog. Since then, the influencer score component associated with the activity of the friend 1 is greater than the influence score component associated with the activity of the friend 2, the news describing the activity of the friend 1 being interacted with can be provided to the user being browsed.

FIG. 1 illustrates a process for providing news 190 to one or more viewing users 120 based on an influencer score. The process begins with an impact targeting engine 175 that identifies the activities 112 performed by the interacting user 110 in the object 130. For example, object 130 may be an article published on a particular web page. The interacting user 110 may have performed an activity on the object by sharing a link to the article. When identifying an activity 112 performed by an interacting user 110, the influence targeting engine 175 identifies one or more visiting users associated with the interacting user 110. In FIG. 1, influence targeting engine 175 identifies browsing users 120a and 120b. When identifying one or more browsing users, the influencer targeting engine 175 determines the influencer score of the user 120 being browsed and the influence of the interacting user 175 on the user 120 being browsed. Influent scores may be associated with categories assigned to activity (112). In one embodiment, the influencer score may be based on the influence of the user 120 being browsed on users who are associated with the user being viewed and on the influencer scores of the users associated with the user being viewed. In Figure 1, for example, the influencer score of the user 120a being browsed may be based on the influence of the user 120a being browsed to the users 122a and 122b and the influencer score of the users 122a and 122b have. Based on the calculated influencer score of the user 120 being browsed and the influence of the interacting user 110, one or more viewing users 120 are selected and the activities performed by the interacting user 110 (190), which is to be described.

In one embodiment, the selection of one or more browsing users 120 may be based on a critical influencer score. In particular, the user 120 being browsed may be selected only if the user's influencer score exceeds a particular threshold influencer score. In this way, the Influent Threshold Score can serve as a hard filter in the distribution of the news. Critical Influencer scores may be set automatically by the social networking system or received from an advertiser or system operator.

In another embodiment, the selection of one or more browsing users 120 may be based on influencer scores and / or other criteria of the users being viewed. For example, when selecting one or more browsing users, the influencer score of the browsing users may be considered along with age, gender, interest, clickthrough rate (CTR) and / or other criteria associated with the users being viewed. This embodiment contemplates the critical score to be one of several inputs in the selection process, as opposed to filtering on-view users based on a critical influence score. For example, an influencer score may be one of several characteristics used to calculate a score for ranking a candidate or advertisement. In this embodiment, the higher the influencer score of the user being browsed, the more likely that the user being browsed is selected to be provided with news 190. However, the news need not satisfy a particular critical influence score to be selected for the user being viewed.

In one embodiment, the viewing user 120 provided with the news 190 may interact with the news so that the news 190 can be presented to one or more users 122 associated with the user being viewed. In this embodiment, the influencer score is determined for the users associated with the user 120 being browsed. The influence targeting engine 175 may then determine which user 122 will be presented with news 190. [

System structure

FIG. 2A is an upper layer block diagram illustrating a system environment suitable for operation of the social networking system 100. FIG. The system environment includes one or more client devices 202, one or more third party websites 203, a social networking system 100, and a network 204. Although only three client devices and one third party web site are shown in FIG. 2A, it should be understood that any number of such entities (including millions) may be included. In alternative arrangements, other entities may be included in the system.

In general, the network 204 may be any network, including but not limited to the Internet, a mobile network, a LAN, a wired or wireless network, a private network, and / or any combination of virtual private networks.

The client device 202 includes one or more computing devices capable of receiving user input and transmitting and receiving data over the network 204. For example, the client device 202 may be a desktop computer, a laptop computer, a tablet computer (pad), a smart phone, a personal digital assistant (PDAs) or any other device that includes computing and data communication capabilities. The client device 202 is configured to communicate via the network 204, which may include any combination of local and / or wide area networks, using both wired and wireless communication systems. The client device 202 may provide a means by which various users may communicate with the social networking system 100. The third party website 203 is connected to the network 204 for communicating with the social networking system 100.

As described herein, the social networking system 100 includes a computing system that allows users to communicate, interact, and access content. In one embodiment, the social networking system 100 may include a user profile that describes users of a social network that include personal information such as career, education, hobbies, preferences, location, etc., demographic information, and other types of descriptive information . The social networking system 100 additionally stores other objects such as fan pages, events, groups, advertisements, general posts, and the like.

FIG. 2B is an exemplary block diagram of various components of the social networking system 100. FIG. The social networking system 100 includes a web server 250, a data logger 260, an impact targeting engine 175, a profile store 205, a group store 210, an event store 215, an application data store 220, A transaction store 225, a personal information data store 230, a relationship data store 240, an activity data store 245, an advertisement store 246 and an influence score store 248. In alternative arrangements, other components may be included in system 100.

Typically, the web server 250 connects the social networking system 100 to one or more third party web sites 203 as well as one or more client devices 202 via the network 204. The web server 250 may include a mail server or other messaging functionality for receiving and routing messages between the social networking system 100 and the client device 202 or third party web site 203 . The message may be an instant message, a queued message (e.g., e-mail), text and an SMS message, or any other suitable messaging technique. In one embodiment, the web server 250 may receive a user request for content, such as news feed news, sponsored news, etc., provided with the content. In response, the web server 250 may send a request for the news to the impact targeting engine 175.

The data logger 260 may receive communications from the web server 250 for other interactions that a user may have in the social networking system 100 a number of different types of social networking objects. The social networking system 100 may manage such data in any suitable manner. In one embodiment, each profile store 205, group store 210, event store 215, application data store 220, transaction store 225, personal information data store 230, The activity data store 245, the ad store 246 and the influencer score store 248 store data structures to store data for each instance of the corresponding type of social networking object managed by the system 100 . The data structure contains an information field suitable for an object of that type. (E.g., the event store 215 includes a data structure that includes the time and location for the event, while the profile store 205 includes a data structure with a field suitable for describing the user's profile) The system 100 starts a new data structure of that type, assigns a unique object identifier to the data structure, and begins to add data to the object as needed. For example, this may occur when a new user activity is detected by the system 100. In response, the system 100 may create a new instance of an activity object in the activity data store 245, assign a unique identifier to the activity object, fill the activity object with information describing the activity or interaction, will be.

Influence targeting engine 175 calculates the influencer score of one or more browsing users and determines based on the influencer score which browsing user the news 190 will be sent to. In one embodiment, the impact targeting engine 175 may identify activities performed by the interacting user 110. For example, the impact targeting engine 175 may search the information collected by the data logger 260 to determine if the interacting user 110 liked a particular movie, liked the fan page, shared a particular link, You can decide whether you shared an article, shared a photo album, and so on. When identifying an activity, the influence targeting engine 175 determines the category for the activity performed by the interacting user 110. A category may include content of an object included in an activity (e.g., an article's topic), metadata about an object included in an activity, identification information about the object (e.g., a universal resource identifier used to connect to an article may be attached) Or any combination thereof. For example, the interacting user 110 may indicate that he or she likes a particular article discussing the sport of football. The influence targeting engine 175 may then process the text of the article to determine the subject (i.e., football) of the article using a natural language processing algorithm. As a result of this determination, the impact targeting engine 175 assigns the activity of the interacting user to the sports or football category.

As another example, the interacting user 110 may indicate that he or she is liking a website on a particular electronics retailer. Based on the tag or other metadata embedded in the web site, the impact targeting engine 175 assigns the activities of the interacting user to the electronic category. In this example, the impact targeting engine 175 may determine the category of activity by referring to an appropriate data structure (e.g., a table) that includes a correspondence between the metadata and the category.

As another example, an interactive user may share an article from a particular financial website. Because the article is selected from the financial website, regardless of the content, the impact targeting engine 175 can assign the activity of the interacting user as a financial category. In this example, the impact targeting engine 175 may determine a category for the activity by referring to an appropriate data structure (e.g., a table) that includes a correspondence between the universal resource identifier (e.g., URLs, .

After identifying the category for the activities performed by the interacting user 110, the influence targeting engine 175 identifies one or more visiting users 120. Each browsing user may be a user associated with the interacting user 110 via the social networking system 100. For example, the user being viewed may be a friend, a follower, a fan, etc. of the user 110 interacting with. After identifying one or more browsing users 120, the influencer score component of the interacting user 110 for each browsing user may be calculated. As described below, the calculated influencer score component can be used to determine which browse user 120 is to transmit the news 190. [ In one embodiment, the influencer score component may be calculated as a function of the influence value of the interacting user 110 for a given viewing user 120 and the influencer score of the user being viewed.

In one embodiment, the impact targeting engine 175 may recursively calculate the influencer score of the user 120 being browsed. In this embodiment, the influence targeting engine 175 may calculate the influencer score of the user's connection relationship 122 being viewed, the influencer score of the set of users associated with the connection relationship 122, and so on. In another embodiment, the influencer score of the user being viewed may be pre-computed. In particular, the influencer targeting engine 175 may periodically calculate and store the influencer scores of the various categories for the users of the social networking system 100 in the influencer score store 248. In this embodiment, the influence targeting engine 175 may retrieve the influencer score of the user 120 being browsed from the influencer score store 248, where each influencer score is associated with the interacting user 110 Related to the activity performed by the user.

In one embodiment, the influencer score of each predetermined visiting user 120 may be calculated based on a set of influential score components of the user 120 being viewed. Each influencer score component of a given viewing user 120 is associated with the influencing user value of the user being viewed on one of the connecting relationships 122 that is weighted against the influencer score of his or her & As shown in FIG.

In one embodiment, each influence value of the user 120 being browsed to one of his or her connection relationships 122 is determined by the content of the user 120 being browsed It can be based on the number and type of interactions. In particular, the impact targeting engine 175 may identify one or more content items associated with the user 120 being browsed. Each content item may be content generated by the user viewing it, news describing the activity performed by the user viewing it, and the like. For example, the content item may be a comment posted by the user 120 browsing to the fan page of a particular company. As another example, the content item may be news indicating that the viewing user 120 liked a particular movie. In one embodiment, the impact targeting engine 175 may identify one or more content items by referring to various stores of the social networking system 100.

When identifying one or more content items, impact targeting engine 175 determines the category for each item. Influence targeting engine 175 may determine the category for the item in any suitable manner. For example, the influence targeting engine 175 may determine whether the information contained in the content item (e.g., text, audio and / or video of the item), metadata associated with the content item For example, URLs). ≪ / RTI >

After determining the category for the content item, the influence targeting engine 175 determines a set of candidate content items. In one embodiment, for each content item, the impact targeting engine 175 determines whether the category assigned to the content item matches the category assigned to the activity performed by the interacting user 110. [ Once a match is determined, the content item is included in the set of candidate content items. In another embodiment, the determination of the influencer score of the user being viewed may not be related to a particular user activity. For example, the influencer score of the user being viewed may be periodically calculated by the influence targeting engine 175 for various categories. In this embodiment, the impact targeting engine 175 determines whether the category to which the Influent score calculated is associated with the category assigned to the item of content.

When identifying a set of candidate content items, the impact targeting engine 175 identifies a subset of the candidate content items, with each item in the subset having a predetermined connection relationship 122 of the user 120 being browsed, It is a content item that worked. For example, the impact targeting engine 175 may include a content item in its subset if the connection relationship 122 likes, shares, reviews, and / or comments the content item.

When identifying a set of candidate content items, impact targeting engine 175 determines the number of interactions performed by connection relationship 122 in the subset of content items. For example, the subset may include two content items. The connection relationship 122 may interact with the first content item twice. The connection relationship 122 may further interact with the second content item three times. Thus, the number of interactions performed by the connection relationship 122 may be five. In addition to determining the number of interactions performed by the connection relationship 122, the impact targeting engine 175 determines the type for each interaction performed by the connection relationship 122. [ Based on the type determined for each interaction, the impact targeting engine 175 assigns a particular weight to the interaction. In one embodiment, the interaction that further indicates the influence of the user 120 being browsed may be weighted higher. For example, the interaction that the connection relationship 122 has commented on for a particular content item may have a greater weight than the interaction where the connection relationship 122 liked a particular content item. In one embodiment, the impact targeting engine 175 also determines the type of relationship for the user 120 being browsed and the connection relationship 122. [ For example, the impact targeting engine 175 may determine that the relationship between the user 120 being browsed and the link relationship 122 is a friend type.

Based on the number of interactions performed by the connection relationship 122, the weight for the interaction, and the relationship type for the user being viewed and the connection relationship, the impact targeting engine 175 determines whether the connection relationship 122 is being viewed The influence value of the user 120 is calculated. The influence value can be calculated in any suitable manner. In one embodiment, the weight of each interaction may be multiplied by a relationship constant (a constant representing the type of relationship between the user and the connection being viewed) after the weights of the respective interactions are summed to obtain an influence value. By way of example, the first interaction may have a weight of one and the second interaction may have a weight of two. The values can be summed and multiplied by a relational constant of one. As a result, the influence value of the user being browsed can be calculated and become a value of 3. In this embodiment, the higher the weight, the greater the number of interactions, and the higher the relationship constant, the higher the influence value calculated by the impact targeting engine 175. Other calculation methods for determining influence values may also be used.

After determining the influence value of each connection relationship 122, the influence targeting engine 175 determines a set of influencer scores of the connection relationship 122. [ In one embodiment, the impact targeting engine 175 may recursively calculate the influential score of the connection relationship 122. [ In another embodiment, the influential score of the connection relationship may be calculated in advance. In particular, the influencer targeting engine 175 may periodically calculate and store the influencer scores of the various categories for the users of the social networking system 100 in the influencer score store 248. In this embodiment, the impact targeting engine 175 may retrieve the influencer score of the predetermined connection relationship 122 from the influencer point store 248. [

When the influence value corresponding to the connection relationship 122 and the influential score are obtained, the influential score of the user 120 being browsed can be calculated. In one embodiment, the set of influencer score components of the user being viewed may be calculated first. Each Influent Score may be a function of the Influence Value corresponding to the same connected relationship 122 weighted against the Influent Score of a particular Connected Relationship 122. [ In one embodiment, the influencer score component may be calculated by multiplying the influence value corresponding to the connection relationship 122 by the influence score of the connection relationship 122. [ After calculating the set of influencer score components, influential targeting engine 175 sums the components to determine the influencer score of user 120 being browsed. In one embodiment, the influencer score of the user 120 being viewed may be stored, for example, in the influencer score store 248. [

As discussed, the influencer score component of the interacting user 110 may additionally be based on the influence value of the interacting user 110 for the identified identified users. The determination of each influence value of the interacting user 110 for a given viewing user may be performed in a manner similar to the determination of the influencing value of the user viewing the connection relationship as described above.

Based on the influencer score of the user 120 being browsed and the influence value of the interacting user 110, the influencer targeting engine 175 may determine the impact of the interacting user 110 on each visiting user 120 Calculate the Influent Score component. When calculating the respective influencer score components of the interacting user 110, the score components can be combined to determine the influencer score of the interacting user 110. The Influencer Score component and / or Influencer score for the interacting user 110 may then be stored, for example, in the Influencer score store 248. In one embodiment, the influencer score component and / or the influencer score may be stored in an appropriate data structure, such as a database. The stored scores / components may then be used to distribute news describing the activities performed by the interacting user (110). The stored score / component may be further referred to later while calculating the influential scorecard elements of the other users who are (directly or indirectly) connected to the interacting user 110.

In one embodiment, the impact targeting engine 175 provides one or more viewing users 190 to describe the user's activities. In one embodiment, such a determination is not made for each browsing user until a request for content containing one or more posts is received from the browsing user. For example, a decision may not be made for a given viewing user until the viewing user requests a page that includes a news feed, sponsorship news, and the like.

The determination of whether the viewing user will be provided with news 190 may be performed in any suitable manner. In one embodiment, the impact targeting engine 175 may compare the impactor score component of the interacting user 110 corresponding to the user being viewed with a predetermined threshold. If the Influent Score component exceeds the threshold, the news may be provided to the viewing user. The threshold may have been previously received, e.g., via an input provided by an operator of the social networking system 100. In another embodiment, the influence targeting engine 175 may compare the influencer score components corresponding to the user being viewed with other influencer score components corresponding to the same viewing user, And other activities performed by the users associated with the user. If the influencer score component associated with the activity of the interacting user 110 exceeds another influencer score component, the news may be provided to the viewing user. For example, the interacting user 1 associated with the user being viewed may like a particular movie. Interactive user 2 associated with the user being viewed can share certain articles posted on the website. Two distinct influencer score components for the user being viewed can be calculated. Influent score components can be compared and the activity associated with the higher influencer score component can be selected to have news describing the activity being presented to the user being browsed. In another embodiment, the influencer targeting engine 175 may consider influencer score components in accordance with other criteria associated with the user being viewed when deciding whether or not the user is to receive the news.

After determining which users are being viewed to receive the news, the impact targeting engine 175 creates a post describing the activities performed by the interacting user 110. For example, the generated news may indicate that the interacting user has liked the fan page of a particular company. In one embodiment, the news may optionally include an input element. The input element allows the viewing user to interact with the post. For example, the viewing user 120 may use the input element to share posts with users who are associated with the user being viewed. The input element may comprise any suitable mechanism for allowing input to be received from the user being viewed. For example, the input element may include a text field, a set of radio buttons, a set of checkboxes, a drop-down menu, a set of buttons or buttons, and the like.

After generating the news, the influential targeting engine 175 sends the news to the browsing user 120 determined to send the news for display. In particular, the client 202 of the user 120 being browsed may be set to display the news to the user viewing it. The generated news may be displayed to each selected browsing user 120 along with an application associated with a social networking system running as a news feed, a banner, an interactive popup, a news feed, or on a client 202.

In one embodiment, after providing the news to the viewing user 120, the social networking system 100 may receive information from the viewing user 120 through the optional input element of the news. In one embodiment, the user's interaction with the input element and the received information can be used to distribute news and / or information about the interaction of the user being viewed to the connection relationships 122 of the user being browsed . For example, the viewing user 120 may share posts containing links posted by the interacting user 110. In response to this sharing, the social networking system 100 may provide news to the connection relationships 122 of the user 120 being browsed. In one embodiment, the news may be provided to the connection relationships 122 based on the influence scores of the connection relationships for display.

Influent  How to provide posts based on your score

Figure 3 illustrates one embodiment of a process for providing news based on an Influent score. In one embodiment, the process identifies an activity performed by the interacting user 110 (315). For example, the process may identify that the interacting user 110 has read a particular news article. The process determines 320 the category of activity performed by the interacting user 110. For example, the interacting user 110 may share a link to an article posted on an investment blog. Thus, the category for the activity may be considered to be finance. The process identifies (325) one or more browsing users associated with the interacting user (110). For example, each of the one or more browsing users may be a friend of the interacting user 110, a user of the interacting user, a fan of the interacting user, and so on. The process determines 330 the influencer score of one or more viewing users 120. Each influencer score may be a function of the influencing value of the user being viewed and the influencer score of the linking relationships for one or more of the linking relationships of the user being viewed. After determining the influencer score, the process determines (335) at least one browse user to be notified based on the calculated influencer score. In one embodiment, the process further determines at least one viewing user based on an interactive user's influence value for at least one viewing user. The process creates (340) a post describing the activities performed by the interacting user (110). The process provides the news to at least one viewing user for display (345).

summary

The foregoing description of embodiments of the present invention has been presented for purposes of illustration and is not intended to be exhaustive or to limit the invention to the precise forms disclosed. Those skilled in the art will appreciate that various modifications and changes may be made thereto without departing from the scope of the present invention.

Some portions of the disclosure describe embodiments of the present invention in terms of algorithms or symbolic representations of operations on information. These algorithmic descriptions or representations are commonly used by those of ordinary skill in the data processing arts to convey the substance of their work effectively to those of ordinary skill in the art. These operations, which are functionally, computationally, or logically described, are understood to be implemented by a computer program or equivalent electrical circuit, microcode, or the like. Also, it is often found that the arrangement of such operations can be referred to as modules without loss of generality. The described operations and their associated modules may be implemented in software, firmware, hardware, or any combination thereof.

Any of the steps, operations, or processes described herein may be performed or implemented by one or more hardware or software modules, or by a combination of these and other devices. In one embodiment, a software module is embodied as a computer program product, including a computer readable medium having computer program code executable by a computer processor to perform some or all of the described steps, operations, or processes do.

Embodiments of the invention may also be associated with apparatus for performing the operations herein. The device may include a general purpose computing device that may be specifically configured for the requested purpose and / or selectively activated or reconfigured by a computer program stored on the computer. Such a computer program may be stored on a computer readable storage medium of any type or in any type of media suitable for storing electronic instructions that may be coupled to a computer system bus. In addition, any computing system referred to herein may include a single processor, or it may be a structure employing a multiprocessor design for increased computing power.

Embodiments of the present invention may also relate to computer data signals included in a carrier wave, wherein the computer data signal may comprise any embodiment of a computer program product or other data combination described herein. The computer data signal is a product that is presented in the type of media or carrier wave and modulated or otherwise encrypted on a carrier wave, type, and transmitted in accordance with any suitable transmission method.

Finally, the language used herein is theoretically selected for purposes of readability and instruction and is not selected to limit or limit the inventive idea. Accordingly, the scope of the present invention is not to be limited by this detailed description, but is limited only by the scope of the appended claims. Accordingly, the disclosure of embodiments of the present invention is intended to be illustrative, and not to limit the scope of the invention. The scope of the invention is set forth in the following claims.

Claims (22)

Identifying an activity performed by an interactive user of the social networking system;
Determining an influencer score of each of the connection relationships in the set of connection relationships of the user being browsed; determining an influence value of the user being browsed for each connection relationship in the set of connection relationships; Determining an influencer score of the user being browsed based on the influencer scores of the set of connection relationships weighted by the influencing user's influencing values for the user being browsed Determining a score;
Determining to transmit, based on the influencer score of the user being browsed, a news describing the activity to a user viewing it; And
Sending a message describing the activity to the client device for display to the user being browsed,
The influence value for each connection relationship indicates the influence of the user in browsing the connection relationship, and the influence value for one connection relation among the sets of connection relations is the content related to the user being browsed and one or more Lt; / RTI > is determined based on interaction. The method according to claim 1,
The step of determining the influencer score of the user being browsed further comprises assigning the category to an activity performed by the interacting user,
Wherein the influence value of the user being browsed and the influential score of the link relationship are each associated with an assigned category. 3. The method of claim 2,
Assigning a category to an activity comprises: based on at least one of content related to a social networking object included in the activity, metadata associated with the activity, and identification information about the activity. The method according to claim 1,
Wherein the influence value of the user being browsed is based at least in part on the content related to the user being browsed and the total count of one or more interactions performed by the linkage. The method according to claim 1,
Wherein the impact value for the connection relationship is based at least in part on the weight for each interaction in the one or more interactions and the weight for the interaction is based on the type of interaction. 6. The method of claim 5,
Wherein the interaction with the comment type has a greater weight than the interaction with the like type. The method according to claim 1,
Further comprising determining an impact value of an interacting user for a user being viewed,
Wherein determining to send the news to the viewing user is based at least in part on the impact value of the interacting user. 8. The method of claim 7,
The step of determining to send the news to the viewing user comprises:
Determining an influencer score component of the interacting user based on the interacting user's influence value and the influencing user's influence score;
Comparing the influencer score component against a threshold value; And
Determining that the influencer score component is greater than a threshold value. Determining, by the computer system, an influence value indicating a first user affects a second user associated with the first user; determining an influencer score of the second user; Generating an influencer score for a first user based on (1) the determined influence value and (2) the determined influencer score; And
Associating the user profile of the first user with the user profile of the first user to store the influencer score of the first user,
Wherein the influence value is determined based on content associated with the first user and one or more interactions performed by the second user. 10. The method of claim 9,
Wherein the second user's influencer score is based on (1) an influential value indicative of the amount by which the second user affects the third user, and (2) the influencer score of the third user. 10. The method of claim 9,
Wherein the influence value is determined based at least in part on a plurality of interactions between the first user and the second user. 10. The method of claim 9,
Wherein the impact value is determined based at least in part on the number of instances in which the second user interacts with the social networking system content associated with the first user. 10. The method of claim 9,
The step of determining the influence value comprises:
Assigning an interaction type to each of the one or more interactions;
Determining, for each of the one or more interactions, a weight based on an interaction type assigned to the interaction; And
Determining an influence value based at least in part on a determined weight for one or more interactions. 10. The method of claim 9,
Identifying an activity performed by a third user associated with the first user;
Selecting a first user to send a post describing an activity performed by the third user based at least on the influencer score generated for the first user; And
And sending the news to the first user. Storing connection relationships for each of the plurality of users of the social networking system;
Determining an influence value of each user of a plurality of users with respect to a user's connection relationship; And
Calculating, by a computer of the social networking system, an influencer score of each user,
Each connection relationship is another user of the plurality of users that the user is influencing,
Each influence value indicates the amount that the user affects the connection relationship,
The influence value on the connection relationship determined for the user is determined based on one or more interactions performed by the content related to the user and the connection relationship,
The influencer score is based on (1) a user's influence value on a user's connection relationship and (2) a set of influential user scores on a user's connection relationship. 16. The method of claim 15,
Each influence value is determined based on at least a part of (1) the total number of the plurality of interactions performed by the user's social networking content and the user's connection relationship, and (2) the interaction type for each of the plurality of interactions. A computer implemented method. 16. The method of claim 15,
Each influencer score of each user is based on the sum of the set of influencer score components, and each influencer score component in the set includes (1) the influential score of the user's connection relationship and (2) the user Based on the value of the influence of the computer. 16. The method of claim 15,
Wherein a computed influencer score of each user is associated with a particular category. 16. The method of claim 15,
Further comprising sending to the user a message describing the user activity based in part on the user's computed influencer score. 20. The method of claim 19,
Determining if the calculated influencer score exceeds a critical influencer score; And
And if the calculated influencer score exceeds a threshold influencer score value, selecting a news to send to the user. 20. The method of claim 19,
Determining a score for the news based on the calculated influencer score; And
Selecting a news to send to the user based at least in part on the score for the news. delete

Images