KR20110127245A - Adheat advertisement model for social network - Google Patents

Abstract

In one implementation, a computer-implemented method includes receiving, at a server, information indicative of activity levels of computer-implemented social network users or acquaintances of the users on the computer-implemented social network. . The method further includes generating an impact score for the users based on the information received by the server. The method also includes transmitting an advertisement from the propagating user to the receiving user when the difference between the first influence score of the propagating user and the second influence score of the receiving user is greater than a threshold value, thereby communicating between acquaintances users. Cyclically propagating the advertisement by a server through the computer-implemented social network.

Description

ADHEAT ADVERTISEMENT MODEL FOR SOCIAL NETWORK}

This disclosure describes methods and systems for placing advertisements on a social network as a whole.

The present disclosure generally relates to providing advertisements to users of a social network. The social network may be an online system that provides a forum for geographically separated users to interact with each other. Social networks can be targeted in other aspects of social interaction, such as friendship and business networking.

Methods of advertising to users of a social network include targeting users of the social network and social networks with particular content. In some methods, the entire social network is targeted to advertisements relating to the general purpose of the social network (eg, business). In other methods, specific users have been targeted for advertisements based on the content of each of the user's profile pages. For example, if a user discusses sports on his / her profile page, an advertisement about sports may target that user.

This disclosure describes techniques for providing advertising to users of a social network. In general, viewers' interactions with particular ads (eg, a user skips the ads) can be monitored, and then the system can display ads that users are likely to see and are less likely to skip. Other advertisements determined as can be selected.

In one implementation, a computer-implemented method includes receiving, at a server, information indicative of activity levels of computer-implemented social network users or acquaintances of the users on the computer-implemented social network. Steps. The method includes generating influence scores for the users based on the information received by the server. The method further includes advertising from the propagating user to the receiving user when the difference between the propagating user's first influence score and the receiving user's second influence score is greater than a threshold. Cyclically propagating the advertisement by the server via the computer-implemented social network among users with whom the acquaintance is related.

In another implementation, a system for distributing advertisements to users of a social network includes one or more servers; An interface to the one or more servers for receiving information indicative of activity levels of social network users or acquaintances of users on the social network; Means for generating influence scores for the user based on the received information; And when the difference between the first influence score of the propagating user and the second influence score of the receiving user is greater than a threshold, sending an advertisement from the propagating user to the receiving user, thereby making the social network between users with acquaintances relevant. And a cyclical propagation module installed in the one or more servers to cyclically propagate the advertisement through.

In another implementation, a computer-implemented method may include outputting an advertisement to a first user of a computer-implemented social network, the output causing the advertisement to be displayed to the first user; Receiving input related to the advertisement from the first user; Retrieving influence scores for the first users, and influence scores for second users of the social network connected with the first user by acquaintance relations; And propagating the advertisement to the second users having an impact score less than the influence score for the first user.

The details of one or more embodiments described herein are set forth in the accompanying drawings and the description below. Other features, aspects, and advantages of the subject matter will become apparent from the description, the drawings, and the claims.

1 is a diagram of an example system for receiving a bid for an opportunity to advertise to users of a social network and propagating advertisements from the user to the user via the social network.
2 is another diagram of an example system for receiving bids for opportunities to advertise to users of a social network and propagating advertisements from user to user via the social network.
3 is a flow chart of an example process for giving a user of a social network an opportunity to display an advertisement based on advertiser requests.
4 is a flow chart of each example process for cyclically propagating advertisements through users of a social network.
5 is a flow chart of each example process for generating a user's interest model from user generated content.
6 is a flow chart of each example process that allows an advertiser to participate in bids for opportunities to display advertisements to users of a social network.
7 is a flow chart of each example process for propagating an advertisement from a propagating user to a receiving user.
8 is a schematic diagram of a computer system.
Like reference symbols in the various drawings indicate like elements.

This disclosure generally describes the placement of advertisements on social networks. More specifically, the specification describes the allocation of opportunities to display advertisements to users and the propagation of advertisements through social networks. In some examples, a method and system are described that allow an advertiser to target users of a social network based on a user's interests and influence on the social network. The opportunity to display ads to an influential user with interests related to the advertisement may be more valuable to the advertiser than the opportunity to display the ad to an influential user with irrelevant interests (or even an influential user with related interests). have. The impact of a user on a social network can be determined by looking at user activity levels and acquaintances in the social network. The advertiser may receive a ranked list of anonymous users according to user interests and influences. The bidding mechanism can be used to accommodate a large number of advertisers seeking to get an opportunity to display an advertisement to a finite and influential number of users on a social network.

In some examples, when an advertiser is given an opportunity to display an advertisement to a particular anonymous user, the advertisement from the advertiser is propagated from that user to the friends of that user using a heat diffusion model. For example, user influence on social networks may be indicated by hit intensity or hit score, with more influential users having high hit scores. The propagation between users can then be modeled using a heat spread model. For example, if a targeted user of an advertisement had a larger "hit" than a user who has not yet been targeted, the advertisement may spread (spread) between two connected users. This allows advertisements to propagate through social networks from more influential users to less influential users. One advantage obtained from the described method is the ability of advertisers to maximize advertising efficiency by disseminating ads from influential users to affected users.

1 is a diagram of an example system 100 for receiving bids for an opportunity to advertise to users of a social network and propagating advertisements from the user to the user via the social network. System 100 may include advertisers 102a-102c, a server 112 that has received a bid, and a social network 118.

Advertiser 102a may submit a request to display advertisement 108 to users of a social network having certain features 106. The server 112 may receive the request and provide the advertiser 102a with a list of anonymous users ranked according to the user impact score and / or user similarity to the requested features 114. The advertiser 102a may then bid on the opportunity to display the advertisement 108 to the ranked one or more users 114. Server 112 may receive bids 104b and 104c from multiple advertisers 102b and 102c for an opportunity to display an advertisement to any anonymous user 114. For example, in the bid chart 114, there are two bids ($ 1.50 for sports ads and $ 0.60 for music ads) for the opportunity to display an advertisement to the user U4. The server 112 then gives the highest bidder an opportunity to display the advertisement, where the advertisement 116 can be displayed on the social network 118.

The advertisement may first be displayed to an anonymously bidding user (eg, user U4 120). The advertisement 116 can then be propagated through the social network 118 according to a connection relationship between users. For example, an advertisement may be propagated from one connected user to another as long as the hit (or impact) score for the propagating user is higher than the hit score for the receiving user. For example, user U4 120 is connected to user U17 122. User U4 has a hit score of 500 and user U17 has a hit score of 450. Since the hit score of U4 is greater than U17, the advertisement may be propagated from user U4 120 to user U17 122. However, since the hit score of U2, which is 510, is greater than the hit score of user U4 (114), the advertisement cannot be propagated from user U4 120 to user U2 126.

2 is another diagram of an example system 200 for receiving a bid for an opportunity to advertise to users of a social network, and disseminating an advertisement from the user to the user via the social network. System 200 may include a client side 202 and a server side 204. The client side 202 can interact with the advertisers 206 to submit requests and bids for displaying advertisements to selected users of the social network. In addition, the client side 202 may interact with the users 208, for example, to display advertisements submitted by the advertiser 206 in a page viewed by the users 208 on the social network. The server side 204 receives requests from the advertiser 206 and, through the bidding process, gives selected users 208 opportunities to display advertisements. The server side 204 may also use, for example, a heat spread model between users 208 of the social network connected by acquaintance relationships (eg, users are designated as friends, users visit other users' profiles, etc.). To propagate ads.

The client side 202 may include an advertiser 206 that submits features of users that the advertiser 206 wants to target for the advertisement. The advertising interface referred to herein as the Adheat interface 210 can receive the features and communicate the features to the server side 204. The server side 204 may include a web server 214 that receives the features, and by the influence of each user in the social network and the similarity to the features submitted by the advertiser through the use of the database 216. Produce a list of ranked users. After ranking, the web server 214 may, in some implementations, forward the ranking list to the client side 202.

In some implementations, the web server 214 can create an ordered list of users in response to receiving the advertisement and / or the content of the advertisement from the advertiser 206. The sorted list may sort the users according to the likelihood that the user will accept the content of the advertisement and / or the amount of influence the user has on other users of the social network. Users listed near the top of the sorted list are more likely to accept the content of the advertisement, and / or are more influential to other users of the social network than users listed near the bottom of the sorted list. Can be.

The ad hit interface 210 at the client side can receive the ranking list of users and provide it to the advertiser 206. The advertiser 206 may provide the ad hit interface 210 with a bid for an opportunity to display an advertisement to selected users on the ranking list. The ad hit interface 210 can forward bids to the server side 204. After receiving the bids, the web server 214 may conduct an auction and, based on the bid, give the ranked users an opportunity to display the advertisement. The auction process is described in more detail below.

The web server 214 can send advertisements to the client side 202. Gadget 212 at client side 202 may display to user 208 an advertisement provided by advertisers 206 that have been awarded an opportunity to be displayed to users 208. The advertisement may then be propagated through the social networks described herein using the friendship graph 218. This is explained in more detail below.

As above, the advertiser 206 may provide the ad hit module 210 with a request targeting the users of the social network as an advertisement target. The request may include information specifying the content of the advertisement and / or features of users that the advertiser wishes to target the advertisement. Features of the user may include demographic information such as age or age range, gender, interests, occupation, income, education, or geographic location of the user.

In some examples, the advertiser 206 may provide the ad hit module 210 with one or more keywords related to the users he wishes to target. For example, the advertiser 206 may provide the ad hit module 210 with a request to display a “sport” related advertisement to “male” users having an “age between 18 and 35 years old”.

In some implementations, the advertiser 206 can select from a drop-down menu descriptors for the content of the advertisement and features of users that the advertiser 206 wants to target the advertisement to. have. For example, drop down boxes may suggest an advertiser with a number of user features from which to choose. In some examples, the advertiser may provide the advertisement itself to the ad hit module 210, where the ad hit module 210 may analyze the advertisement and determine the content of the advertisement. The advertisement provided to the ad hit module 210 may be text, an image, a video, or a combination thereof.

The ad hit module 210 may determine the content of the advertisement using a content modeling method. The modeling method derives the content of the advertisement based on words or symbols used in and / or associated with the advertisement. do. In some examples, the content of the advertisement may be derived from the text of the advertisement. When text or video is provided with an advertisement, optical character recognition can be used to create text for the advertisement. In some examples, the content of an advertisement may be derived from meta information associated with the advertisement.

The ad hit module 210 can submit a request from the advertiser 206 to the web server 214. In response, web server 214 may provide ad hit module 210 with a ranked list of anonymized users that match one or more features specified by advertiser 206.

The opportunity to display an advertisement to users ranked at the top of the list may be more valuable to the advertiser 206 than the opportunity to advertise to users ranked at the bottom of the list. Users ranked at the top of the list may be more influential on the social network and may better fit the features specified by the advertiser 206. The ranking list of users may include information about each user, such as anonymized user identifier (preventing the advertiser from identifying a particular user on a social network), influence score, and relevance score. Users can be ranked on the list according to influence scores, relevance scores, or a combination thereof. In general, the larger the user influence and relevance scores, the more likely the user is to appear above the ranking list.

The impact score may be a metric of user influence over other users of the social network. Influence scores may be based on activity levels and / or acquaintances of the user. The activity level of the user may be a measure of the user's interaction on the social network.

The acquaintance may be a connection (eg, a friend) between two users on a social network. The influence score of the user may be influenced by the amount and / or importance of the user's acquaintances.

For example, a user's activity level may be determined by the quality and / or amount of interaction between that user and other users in a social network; Can be determined by the quality and / or quantity of content generated by a user in a social network; And / or the frequency of user interaction and / or content generation during the time period in the social network. A user who is good in quality and / or quantity of interacting and / or generated content may be determined to have a greater activity level than a user who is low in quality and / or quantity of interacting and / or generated content. In addition, a user with a high frequency of content and / or interactions generated within a time period may be determined to have a greater activity level.

User interaction in social networks includes at least posting comments to other users' pages, sending messages to other users, participating in forums on social networks, adding other people as friends, and chatting rooms on social networks ( posting a comment in a chat room and / or viewing another user's profile page. Content generated by the user may include uploading at least an image or video, updating a user profile page, posting input to a blog, and / or updating user status information.

The quality of interaction for a user in a social network can be determined by evaluating the substantiveness of the interaction and / or the responsiveness of the interaction, in the context in which the interaction occurred. For example, two comments about sports can be posted as follows: 1) “I like sports” and 2) “I love watching sports, especially football and hockey – at my super bowl party, Check out the photo you posted on my profile. ”Since the latter is more realistic, the latter can be determined to have better quality.

In addition to the previous example, if the two comments appeared in response to a discussion about the genre of music, then the comments are unlikely to have a high level of quality, the comments may be in the context in which they appeared (eg, the genre of music). Because it is not a response. Though not realistic, comments that respond to the context in which they appear may be determined to be realistic but of higher quality than off-topic comments. For example, a comment posted in the discussion on the genre of music, “Blues is my favorite music type,” can be determined to have a better quality than the very realistic second comment described above.

Like the quality of user interaction, the quality of user-generated content may be evaluated for contextuality and responsiveness. In addition, the quality of user generated content can be evaluated based on originality. User-generated content that is original (eg, not already posted on a social network by another user) may be determined to have a higher level of quality than content that is not original (eg, copied from another user's profile page). Can be.

The amount of user interaction and user generated content in the social network may be determined by the number of interactions and the amount of content posted to the social network. For example, a user who posted five comments may be determined to have a greater amount of activity than a user who posted two comments.

In some examples, the activity level of the user may be determined by a combination of user interaction and quality and quantity of user generated content. The activity level may be determined by the weighted amount of user interaction and / or the weighted amount of content generated by the user, with each interaction and / or user generated content being weighted according to its quality. For example, a user who posts two comments similar to the second comment may have a higher activity level than a user who posted five comments similar to the first comment.

In addition, the activity level of the user may be based at least in part on the frequency of user-generated content and / or interactions within the time period. For example, a user who recently interacted with 5 other users in a month, and posted 10 new photos (eg, content creation), interacted with only 2 other users in the last month and only 5 new photos It may be determined to have a higher activity level than other users who have posted. Interactions and / or generated content may be further weighted depending on the quality determined.

The influence score of a user may be influenced by the amount of acquaintance of the user. A user with more friends may be more influential than a user with fewer friends. As such, a user with a greater number of acquaintances may result in a greater influence score than a user with a smaller number of acquaintances.

The influence score of the user can be influenced by the importance of the user's acquaintances. Users with more important acquaintances may have greater influence in social networks than users with less important acquaintances. The importance of a user's acquaintance may be based on the influence score of the user friend (the user connected to that user by acquaintance) and / or whether the acquaintance connects the individual sub-graphs of the social network.

In some examples, acquaintances with users with high impact scores may be more important than acquaintances with users with low impact scores. In another example, an acquaintance with a user with a low impact score may be more important than an acquaintance with a user with a high impact score.

The social network of users may be represented by a graph of users (nodes) connected to each other by an acquaintance relationship (edges), as illustrated by the friendship graph 218. Friendship may be explicit relationships between users of a social network. For example, friend assignments agreed between two users may be explicit. Also, acquaintances can be implicit. For example, although user U3 and U4 do not have an explicit friendship, user U3 visits the user U4 profile page, user U3 interacts with user U4 via an interface on the social network (eg, chat rooms, emails). Exchange, etc.), and / or if user U3 and user U4 share a friend in common (eg, both of them have an explicit friend relationship with user U5), an acquaintance may be implicit between them.

Within a social network graph, a sub-graph is a group of users who have a lot of acquaintances with each other in the group but have little acquaintance with users other than that group. The acquaintances connecting users of two separate sub-graphs may be more important than the acquaintances connecting users in the same sub-graph. For example, users U2, U4, U6 and U17 in social network 118 are connected via acquaintances within the same sub-graph. The connection between the user U10 and the user U4 belonging to a sub-graph separate from the sub-graph to which U4 belongs may be regarded as connecting the users of the individual sub-graphs.

The relevance score may be a unit of similarity between the user's characteristics and the user's characteristics requested by the advertiser 206 as a target of the advertisement. As described above in connection with the advertiser 206 request, the characteristics of the user may include demographic information such as age or age range, gender, interests, occupation, income, education, or geographic location of the user. A user with features similar to the features requested by the advertiser 206 has a greater relevance score than a user with less similar features. For example, assume that user U3 is "male, age 25, likes hockey" and user U5 is "male, age 37, likes blues music." For advertisers who have requested to display advertisements to "male" users, "ages 18-35" who like "sports," user U3 will have a greater relevance score for that request than user U5.

In some examples, the user activity level score may be determined taking into account the relevance of each interaction to the advertiser's 206 request. In such an example, each interaction may be weighted according to the relevance to the features requested by the advertiser 206. For example, using the scenario in the previous paragraph, if user U3 and U5 post the same number of realistic comments on a social network, but most of user U5's posts are about sports, then user U5 is weighted more than user U3. Will have an activity level score.

In some embodiments, web server 214 may receive a ranked list of users by querying database 216 hosted at server side 204. Database 216 may include anonymized information related to users of the social network. For example, the anonymized information may include information related to the user's characteristics, interactions in the social network, influence in the social network, and / or relevance to the advertiser 206 request.

In some examples, database 216 may be populated with models related to users of the social network. The models may provide a synopis of user features (including interests) in a format that is maintained and queried. The models may generate information for insertion into the database 216 by the web server 214 or other server side 204 component not shown. For example, latent topic modeling can be used to generate models. Potential topic modeling can produce models by processing data from a social network (eg, user profile page, user interaction) to obtain user characteristics (eg, interests).

Upon receiving the ranking list of users from the web server 214, the ad hit module 210 can provide the advertiser 206 with the ranking list of users. The advertiser 206 may then provide the ad hit module 210 with a bid for an opportunity to display an advertisement to one or more users included in the ranking list of users. The bid may specify a minimum bid per user, a maximum bid per user, and / or a total budget for propagating the advertisement through a social network. The advertiser 206 may bid for individual users and / or blocks of users in the ranking list.

The opportunity to display an advertisement to higher ranking users in the ranking list may be more expensive than the opportunity to display an advertisement to lower ranking users. It can be more expensive if a user ranked at the top of the list is more influential on the social network and more related to features submitted by advertiser 206 in relation to the advertisement. An advertiser 206 who wishes to win an opportunity to display an ad to a user ranked at the top of the list is likely to offer a higher bid for that user than to lower ranked users. This is because there is a possibility of bidding on the same influential user.

In some implementations, the ad hit module 210 can perform all or some of the functions that the web server 214 performs with respect to the ranked users of the social network. For example, in response to receiving an advertisement request from the advertiser 206, the ad hit module 210 does not send information about the advertisement content of the advertiser 206 to the web server 214 for a user list request. You can send In response, web server 214 may provide a list of users and their related information (eg, user impact scores, user interests) to adhit module 210. The adhit module 210 may then calculate a ranking for the users based on the influence score of the received user and / or the similarity between the received user's interests and the content of the advertisement. After computing the rank list, the ad hit module 210 may display the rank list to the advertiser 206 and receive user bids from the advertiser 206.

In some implementations, the ad hit module 210 can request users that fall within a predetermined influence score range or influence score percentage from the web server 214. For example, the ad hit module 210 may request that the web server 214 return users with an influence score greater than 450. In another example, the ad hit module 210 can request that the web server 214 return users with an impact score within the top 10% of the impact score. As described above, the ad hit module 210 may later rank these users.

The ad hit module 210 can submit a bid from the advertiser 206 to the web server 214. The web server 214 may receive bids for the opportunity to display an advertisement to users of a social network, from the plurality of advertisers 206 via the plurality of ad hit modules 210 (eg, FIG. 1). 104a-c). The web server 214 may conduct an auction for the opportunity to display an advertisement to users of the social network based on the bids received from the advertisers 206. The auction may give the advertiser with the highest bid the opportunity to display an advertisement to the user of the social network.

For example, advertiser A1 may submit a maximum bid of $ 1.5 for the opportunity to display sports-related advertising to anonymous user U1. Advertiser A2 may submit a maximum bid of $ 1.0 for the same opportunity to display an advertisement to user U1, but advertiser A2's advertisement is about music. The auction conducted by the web server 214 will give the user U1 the opportunity to display an advertisement, and the advertiser A1 with the maximum bid.

The web server 214 can propagate the advertisement to the gadget 212 associated with the user 208 after the auction gives the user an opportunity to display the advertisement. In some examples, gadget 212 is an interface that displays advertisements on a page viewable by users in a social network. The advertisement of the winning advertiser 206 may be sent to the gadget 212 (the gadget 212 may display the advertisement to the user 208 via a computer display). Gadget 212 can provide the user with an interface that allows the user to interact with the advertisement. Gadget 212 may be a simple HTML or JavaScript application that may be embedded in a web page or other application. In some examples, user interaction with an advertisement may include a user's own choice of whether or not to propagate the advertisement to other users associated with it by an acquaintance.

In some examples, users who interact with the advertisement may share the revenue of the advertisement. In some examples, the revenue shared with the user is a portion of the revenue due to the user's interaction; Fixed amount; An amount based on the number of users viewing an advertisement and / or interacting with the advertisement due to the user's interaction; And / or an amount based on the characteristics and / or influence of users viewing an advertisement and / or interacting with the advertisement due to the user's interaction. For example, a user who has interacted with an advertisement and decides to propagate the advertisement to other users of the social network may receive a share of revenue for that advertisement based on the number of users that the advertisement has propagated.

In some implementations, the propagation of the advertisement is due to the user only if the user has propagated the advertisement directly through one of the acquaintances. In another implementation, the propagation of an advertisement is due to the user only when possible, which is to trace back the user's interactions in the chain of propagation of the advertisement from the user to the user. In such an implementation, the revenues due to the propagating user may be reduced as the user's propagation gets further away. For example, user U1 may receive 10% of the revenue for disseminating an advertisement to user U2. However, when user U2 propagates the advertisement to user U3, user U1 can only receive 5% of the revenue. Similarly, when user U3 propagates the advertisement to user U4, user U1 may only receive 2.5% of the revenue.

Gadget 212 may propagate the advertisement to other users of the social network based on the hit spread model. Heat diffusion is a concept from thermodynamics. In general terms, heat diffusion defines that heat spreads (spreads) from one adjacent particle to another when the heat (heat) between the two particles is different. For example, if particles A and B are adjacent and they have a temperature of 70 degrees and 50 degrees, respectively, the heat will move from particle A to particle B. However, the heat will not migrate from particle B to particle A. If both particles A and B had a temperature of 70 degrees, no heat transfer would occur.

Using the hit spread model, advertisements are propagated through social networks by assigning a “temperature” to each user. The advertisement is delivered from the propagating user to the receiving user if the propagating user has a higher “temperature” than the receiving user. In some implementations, the heat spread model uses a heat index score to indicate the user's temperature. The hit index score of the user may be based on the influence score (described above) of the user in the social network. The hit index score and influence score of the user may have a direct relationship (eg, a user with greater influence will have a larger hit index score).

Under the heat spread model structure, advertisements can be propagated between “neighboring” users. Two users may be considered adjacent if they are connected to each other in a social network, as indicated by the connections between nodes on friend relationship graph 218. Connections between users can take the form of acquaintances. The acquaintance may be, in particular, a relationship in which users mean friends, colleagues, classmates and / or members of the same group. In another implementation, acquaintances may refer to users viewing a profile with each other, posting content on a profile page for the user, and / or receiving an electronic communication (eg, a chat request from another user). have.

In some implementations, the advertisement is propagated from the propagating user to all users connected to the propagating user and having a lower hit index score than the propagating user. For example, user U3 is a propagating user with a hit index score of 70. User U3 is connected to users U5, U8, and U9 with hit index scores of 50, 35, and 80, respectively. The advertisement displayed to the propagating user U3 will be spread to the users U5 and U8 because they have lower hit index scores. Since user U9 had a higher hit index score than user U3 propagating, the advertisement would not be propagated to user U9. However, when an advertisement is displayed to user U9 and serves as a user to which U9 propagates, the advertisement displayed to user U9 will be propagated to user U3 because user U3 has a lower hit index score.

Under the hit spread model, advertisements are propagated to users of social networks using circulation. The advertisement spreads from the propagating user to the receiving users, each receiving user being connected to the propagating user and having a lower hit index score than the propagating user. After the advertisement is displayed, each receiving user can act as a propagating user. As such, the advertisement is spread from the receiving user to other users with a lower hit index score that are connected to the receiving user.

For example, suppose using the above scenario, user U9 started out as a propagating user. The advertisement is displayed to user U9 and then propagated to users who have a lower hit index score and are connected to him. In this case, because the user U3 has a lower hit index score than the user U9, the advertisement is propagated to the user U3. User U3 then serves as the user to propagate. By U3 as the propagating user, the advertisement is propagated to users U5 and U8 because users U5 and U8 have lower hit index scores than user U3. Thereafter, users U5 and U8 will serve as propagating users. The advertisement will be linked to user U5 and also propagated to users with a hit index score lower than U5. The advertisement will also be linked to user U8 and also propagated to users with a hit index score lower than U8.

Circular propagation of the advertisement may continue until the termination condition is met. In some examples, the termination condition may be when the advertising budget is exhausted. In another example, the termination condition may be that the propagating user chooses not to interact with the advertisement. In such an example, the gadget 212 can replace the advertisement with a replacement advertisement for display to the user 208. In another example, the termination condition may be a hit index score of the receiving user that is greater than the propagating user's hit index score. In another example, the termination condition may be a receiving user who has already received the advertisement.

Gadget 212 may send data about user 208 interacting with the advertisement to web server 214. The transmitted data may include information about viewing and clicking of the advertisement.

The web server 214 may consist of one or more servers. The web server 214 may include an interface 220, a request scoring module 222, an advertisement assignment module 224, and a circular propagation module 226. The interface 220 can communicate with the client side 202. The request scoring module 222 may score a bid received from the advertiser 206 for an opportunity to display an advertisement to at least one user of the social network. The ad assignment module 224 can assign an opportunity to display an advertisement to a user of the social network based on the bids scored by the request scoring module 222. The recursive propagation module 226 may recursively propagate the advertisement to users of the social network based on the user impact scores.

3 is a flow chart of an example process 300 that grants users of a social network an opportunity to display an advertisement based on an advertiser request. Processes 300 may be performed by a system such as, for example, systems 100 and 200, and for the sake of clarity of representation, the description below uses system 100 and 200 as an example to describe processes. . However, other systems or combinations of systems may be used to perform the process 300.

Process 300 may begin at step 302 by receiving an advertiser's request to target users of a social network with specified features to target an advertisement. The advertiser's request may be received at the server side from the client side. For example, the web server 214 may receive a request from the advertiser 206 at the client side 202.

In step 304, process 300 may rank the users of the social network based on the similarity and user influence scores of the user features to the specified features. Ranking the users may be performed in response to a request received from the advertiser. For example, web server 214 may classify users based on profiled information and information submitted by advertiser 206, which information specifies the characteristics of the users that the advertiser wishes to target. In some implementations, the web server 214 can presort or rank users based on the impact. For example, the influence rank of a user may be updated on a period basis (eg, every hour). Next, if users are ranked according to features submitted by the advertiser, the pre-calculated ranks for influence may be combined with feature ranks to generate a user rank based on user influence and features.

In some embodiments, process 300 may also include sending a ranked list of users to the advertiser making the request. Some embodiments may also include receiving advertiser bids for opportunities to display advertisements to users on the ranking list. Advertiser bidding may include bidding on an opportunity to display an advertisement to an individual user and / or bidding on an opportunity to display an advertisement to a group of users. For example, an advertiser may bid $ 1.00 for an opportunity to display an advertisement to users U3 and bid $ 2.50 for an opportunity to display an advertisement to users U4-U8.

In some implementations, the group of users can be a range of ranked users. For example, an advertiser may bid for an opportunity to display an advertisement to users ranked first to tenth on the ranking list. In some implementations, the bid for a group of users may correspond to an amount (eg, $ 1.00 for all users U4-U8) for all users in the group. In another implementation, the bid amount for a group of users may correspond to the amount for each user in the group (eg, $ 1 for each of users U4-U8).

Process 300 may score advertiser bids for an opportunity to display an advertisement to users of the social network at step 306. For example, the scoring may be based on the amount the advertiser bids for the opportunity to display the advertisement to the user of the social network. There may be a direct relationship between the resulting score and the bid amount (eg, the larger the bid amount, the larger the resulting score).

The bids may also be scored based on how relevant users have found advertisers' ads in the past. For example, the database at server side 204 may store information indicating historical click-through rates for advertisements submitted by particular advertisers. Bids from advertisers with ads with high click-through rates may be scored higher than bids from advertisers with ads with low click-through rates.

In some implementations, one score may be provided for an advertiser bid for an opportunity to display an advertisement to a group of users. In some implementations, advertiser bids for opportunities to display advertisements in a group of users may be provided with a score for each individual user of that group. In some embodiments, the score for an individual user of a group may be the same as the scores for other users of the group (eg, score for an individual user = score for the user groups / number of users). In another embodiment, the score for an individual user of the group may be the quotient of the score for the group, which is proportional to the influence score of that user relative to the influence score for other users (eg, individual user score = for the user group). Score × (individual user's influence score / group's total influence score).

In step 308, an advertising opportunity may be assigned to the advertiser. Allocating an advertising opportunity may be based on a score of advertiser bids and / or a score associated with users whose advertisement is to be displayed. Advertisers may be assigned advertising opportunities for individual users and / or group users. The bids considered when allocating an advertising opportunity may be limited to bids received within a certain time range and / or bids for displaying an advertisement to a user within a specific date range.

In some implementations in which an advertiser submits a group bid and group bids are scored for each user in the group, the opportunity to display an ad in the group is determined by advertisers with different bid scores for each user in the group. If greater than the bid scores, it may be assigned to that advertiser. For example, if an advertiser bids on displaying an ad to users U4-U8 and the bid is scored individually for each user (eg, 1.0 for each user), some of the users U4-U8 If no advertiser has a bid score greater than 1.0 (or smaller depending on the scoring technique) for, the advertiser may be given the opportunity to display the advertisement to the group.

In some other implementations in which an advertiser submits a group bid and group bids are scored for each user in the group, the opportunity to display an advertisement to each user in the group may be evaluated and assigned individually for each user. For example, if an advertiser bids on displaying an advertisement to U4-U8, and the bid is scored individually for each user (eg, 1.0 for each user), the bid is greater than 1.0 for user U4. If no advertiser has a score, the opportunity to display an advertisement to user U4 may be assigned to that advertiser. The opportunity to display an advertisement to users U5-U8 may be assigned in a similar manner, independent of evaluating and assigning the opportunity to display to other users in the group.

In some implementations in which an advertiser submits a group bid and the group score is provided to the group bid, the opportunity to display in the group may be assigned to the advertiser with the highest scored bid for that group. For example, if the first and second advertisers had bid scores of 2.5 and 2.7 for the opportunity to display the ad to users U4-U8, then the second advertiser would have the opportunity to display the ad to users U4-U8. Can be assigned.

In some implementations in which an advertiser submits a group bid and the group bid is provided to the group bid, the minimum score required to display the advertisement to each individual user is determined, the minimum score for the group is aggregated, and the combined minimum score. Can be assigned by assigning an opportunity to that advertiser if is less than or equal to the group bid score. For example, if an advertiser's bid score for a group of users U4-U8 is 2.5 and the overall minimum score for displaying an advertisement in that group is 2.4 (eg, the minimum for displaying an advertisement for each of the users U4-U8). Scores are 1.0, 0.8, 0.2, 0.2 and 0.2, respectively, and the advertiser may be assigned an opportunity to display in that group. The minimum score requested to display an advertisement to the user may be determined by identifying the highest bid score for the user and increasing the highest bid score by a predetermined amount and / or percentage.

In some implementations, there may be more than one opportunity to display an advertisement to a user within a time period. These opportunities may be assigned to the same or different advertisers depending on bid scores. For example, if a user has three opportunities to display an advertisement within a time period, the bids with the three highest scores may be assigned to those three opportunities. For example, if four advertisers A1-A4 bid on an opportunity to display an advertisement to four users U1-U4 and there are three opportunities available per user, first to third place for the user Advertisers who bid will be given the opportunity to display the advertisement to the user. A1 bids $ 2 for users U1-U4, A2 bids $ 1.50 for users U1-U3, A3 bids $ 1 for users U2-U4, and A4 bids users U1 If you bid $ 0.5 for U4, the advertising opportunity may be assigned to each user as follows: U1- A1, A2, A4; U2-A1, A2, A3; U3-A1, A2, A3; U4-A1, A3, A4; U5-No opportunity assigned.

In some implementations, when a user has multiple opportunities to display an advertisement, if a single advertiser has bid on more than one of the multiple opportunities, the single advertiser may be assigned more than one of the multiple opportunities.

After assigning an advertisement opportunity for the user or users to the advertiser, the advertiser's advertisement may be output to the user or users, at block 310. The user or users who receive the advertisement may be on the client side, such as users 118 or user 208 specified in the social network.

In some implementations, users of a social network that are not assigned an advertising opportunity, such as user U5 in the previous example, may receive default ads instead of ads assigned to the advertiser (eg, the default ads are ads of the social network). , Profiles of other users on social networks, advertisements for nonprofits / charities, etc.).

4 is a flow chart of each example process 400 for cyclically propagating an advertisement through users of a social network. For example, processes 400 may be performed by a system such as systems 100 and 200, and the description below uses the process as systems 100 and 200 as an example for clarity of presentation. However, other systems, combinations of systems, can be used to perform the process 400.

Process 400 can begin at step 402 and / or 404. In step 402, process 400 may determine an acquaintance (eg, a friend relationship) between users of the social network. In step 404, process 400 may determine activity levels (eg, number of posts, add friend, etc.) for users of the social network.

In step 406, process 400 may generate an impact score for users of the social network using the determined user acquaintance and the determined user activity levels. Combining the user's acquaintances and activity levels to generate an influence score can be done differently depending on the configuration of the social network and / or advertiser preferences. For example, the first social network may be configured to weight acquaintances twice as much as the activity level, while the second social network may be configured to equally weight acquaintances and activity levels.

In some implementations, the social network can have multiple configurations used to generate impact scores for users of the social network. The social network decision on which of the plurality of configurations to use may depend on the conditions within the social network and / or the type of advertisement to be displayed. For example, the configuration used by the social network during the week may be different from the configuration used during the weekend. Other configurations may be determined by the social network based on demographic information of users accessing the social network for various time periods.

In some implementations, the social network may follow the advertiser's weight for acquaintances and activity levels when generating influence scores for users of the social network. The use of influence scores generated according to an advertiser's weight may be limited to ads from that advertiser. An advertiser can provide the social network weights that apply to particular ads from that advertiser.

Generation of influence scores for users of the social network at step 406 may be performed regularly and / or when certain events occur. For example, influence scores may be generated every 10 minutes, 30 minutes, 1 hour, 6 hours, 12 hours, day, 3 days, week, and / or month. As another example, impact scores may be generated when a certain number of activity events occur within a social network (eg, generate an impact score each time: 1000 comments are posted on a social network, 10 users are social Joining a network, creating 100 contacts, etc.)

In some implementations, generation of influence scores for users of the social network can be triggered by advertisement display for users of the social network at step 406. Whenever an advertisement is first displayed to a user on a social network, the social network may generate an influence score for the users of that social network. When an advertisement triggers generation, the generation of influence scores may be limited to users in the sub-graph of the user who was the initial target of the advertisement.

In some implementations, process 400 can display an advertisement to a user of a social network. The opportunity to display the advertisement to the user may be granted to the advertiser who has won the opportunity through the bidding process, as described, for example, with reference to FIG. 3.

In step 408, process 400 may determine whether the impact score for the propagating user is greater than the influence score for the receiving user. The propagating user may be a user whose advertisement is displayed. The receiving user may be a user connected through an acquaintance (eg, a friend relationship) to the user who propagates. If the impact score for the propagating user is greater than the influence score for the receiving user, process 400 propagates the advertisement to the receiving user from the propagating user, at step 410.

In some examples, steps 402-406 regarding the generation of the influence score may not be performed until step 408. In such an example, steps 402-406 may be performed only for the propagating user and the receiving user. For example, after the advertisement is displayed to the propagating user, process 400 may determine impact scores for the propagating user and the receiving users.

In step 410, propagating the advertisement from the propagating user to the receiving user may include displaying the advertisement to the receiving user. The advertisement displayed to the receiving user may be the same advertisement displayed to the propagating user. After propagating the advertisement to the receiving user at step 410, process 400 may return to step 408. However, returning to step 408, the receiving user may serve as a propagating user, and a user connected through a relationship with the receiving user may serve as the receiving user.

For example, if user U1 is the first propagating user, user U2 is the first receiving user, and the influence score of U1 is greater than the influence score of U2, then the advertisement displayed in U1 may be propagated to U2. After the advertisement has been propagated to U2, process 400 may return to step 408, where U2 serves as a propagating user, and U3, which is connected through an acquaintance with U2, serves as a receiving user.

In steps 408 and 410 the advertisement may cyclically propagate between users of the social network. Steps 408 and 410 may be implemented for each acquaintance relationship of the propagating user. For example, an advertisement displayed to a propagating user may be propagated to each person in acquaintance with the propagating user having an impact score less than the propagating user. Next, each user who receives an advertisement from the propagating user may propagate the advertisement to each user who has a smaller impact score and who is in an acquaintance.

The cyclical propagation of the advertisement at steps 408 and 410 may continue until process 400 determines that the influence score of the propagating user is not greater than the influence score of the receiving user. If such a determination is made, process 400 may end. The cyclical propagation may also end when a condition is met where continuous propagation of the advertisement is not allowed (eg, the conditions outlined with respect to FIG. 7).

In some examples, propagation using a heat spread model may be limited based on the degree of separation between users. For example, if an advertisement propagates from a first user to a user greater than separation 6, propagation may stop regardless of the hit index scores of adjacent users.

5 is a flow chart of an example process 500 for generating a user's interest model from user generated content. For example, processes 500 may be implemented by a system such as systems 100 and 200, and for clarity of representation, the following description uses systems 100 and 200 as examples to illustrate the process. . However, other systems and combinations of systems may be used to implement process 500.

Process 500 may begin at step 502 by collecting information about a user based on user-generated content. The information collected may include demographic information (eg, age, gender, geographic location, etc.) and / or user interests (eg, sports, music, etc.). User-generated content may include information that the user provided on his or her profile page (e.g. interest, age, gender, etc.), groups the user subscribed to (e.g. hiking club group), and / or comments posted by the user (e.g. , Status updates, comments posted on other users' profile pages), and the like.

In some examples, visual and audio media (eg, music, photos, videos) posted by the user may also be collected. Meta-tags associated with visual and audio content may be collected. In some examples, an audio / visual content analyzer may be used to derive information related to visual and audio content. For example, the audio / visual analyzer can identify the audio file by comparing the file signature calculated for the audio file to a database of well known audio files. In another example, the audio / visual analyzer may attempt to find out information about the content of the picture (eg, baseball cap, baseball glove) by looking for items and / or logos in the picture that may be associated with a particular subject. , And / or finding a baseball team logo on an image can represent a photo about baseball).

At step 504, process 500 may enter the collected information into a machine learning algorithm. The machine learning algorithm may be a latent topic modeling algorithm. The latent topic modeling algorithm may examine the collected information based on the presented topic, and generate a topic list appropriate for the user. In some implementations, the algorithm can collate collected information about a database of keywords associated with the topics. The algorithm can dynamically learn and adjust these keywords and associated topics.

At step 506, process 500 may generate a topic model that provides a user's interest based on the collected information. The topic model for the user may provide a list of attributes associated with the user's interests. The topic model can provide an indication of how interested the user is in a particular topic (eg, the user is most interested in sports and then interested in music). After the topic model is generated, process 500 may end.

In generating a topic model, previous topic models may be examined to determine a change in interest of the user over time. For example, the topic model may provide that the user has recently become interested in genealogy and that the user's interest in sports has decreased over the past year.

For example, user U10 may have a profile page with blog entries about football and skiing and pictures of recent professional football matches that the user attended. User U10 may also be a member of a video game group on a social network, and post comments on other users' profile pages about snow conditions in soccer, movie and ski areas. All of this information can be collected by process 500 and entered into machine learning algorithms. Using machine learning algorithms, process 500 can then generate a topic model for user U10. The topic model may list the interests described below, in order from maximum interest to minimum interest: soccer, skiing, movies, and video games. By examining previous topic models of user U10, the topic model can provide that football is consistently a top concern of user U10, skiing is a periodic concern that is a peak in winter, and movies and video games have declined over the past year. .

6 is a flow chart of an example process 600 in which an advertiser bids for an opportunity to display an advertisement to users of a social network. Process 600 may be implemented by a system such as, for example, systems 100 and 200, and for clarity of representation, the following description uses system 100 and 200 as an example to describe processes. However, other systems or combinations of systems can also be used to implement the process 600.

Process 600 may begin at step 602 by the advertiser entering parameters describing characteristics of users that the advertiser wishes to target the advertisement. Features may include demographic information (eg, age, gender, geographic information) and / or interests (eg, sports, music). The advertiser may enter this information through a client-side interface, such as the ad hit interface 210 of the system 200, which information may be submitted to a server side, such as the web server 214 of the system 200. Can be.

In step 604, process 600 receives a list of ranked users from the server, the ranking based on the similarity of the features and the user impact score. The ranking list of users may be generated at a server, such as web server 214, in a manner similar to that described above with respect to FIGS. The ranking list of users may be received via an interface, such as the ad hit interface 210 of the system 200.

For example, advertiser A1 may enter parameters for users to be targeted to an advertisement, such as a male between 18 and 35 years of age interested in sports. Advertiser A1 may then receive a list of ranked users U1-U5. The user U1, which is at the top of the list, can most closely match the features submitted by advertiser A1 and can have a high impact score in the social network. The user U5 located below the list may have the least match to features submitted by advertiser A1 and / or have a low impact score in the social network.

In step 606, the advertiser may enter the maximum and minimum costs that he may pay per user to display the advertisement. In some examples, an advertiser may enter a maximum and minimum cost that he may pay to display advertisements in a block or group of users. The maximum and minimum costs may be entered through an interface, such as the ad hit interface 210 of the system 200.

Using the previous example, the maximum and minimum costs for users U1-U5 that advertiser A1 can pay may be directly related to the ranking of each user in the list received in step 604. Advertiser A1 may enter a maximum and minimum cost of $ 1 and $ 0.5 for user U1, respectively. However, advertiser A1 may enter a maximum and minimum cost of $ 0.25 and $ 0.05 for user U5, respectively. In some examples, advertiser A1 may not enter bids for users of lower rank, such as user U5.

In step 608, the advertiser may enter a total budget for displaying the advertisement. The total budget may be the maximum amount an advertiser can spend in order for an advertisement to be displayed to users in accordance with the bid and propagated to other users of the social network. When the budget is exhausted, the advertisement may stop being displayed to users of the social network. The total budget may be entered through an interface, such as the ad hit interface 210 of the system 200.

In some examples, the budget may exist for a periodic or specific time period. For example, an advertiser may stipulate that the daily budget for displaying an advertisement is $ 50.00. When the daily budget is exhausted, the advertisement may not be displayed to users of the social network until the next day begins.

In step 610, the advertiser may submit bids (eg, maximum and minimum cost) for an opportunity to display an advertisement to users from the ranking list. In addition to bidding, the advertiser can also submit a total budget for the ad. The bid may be submitted by the advertiser via an interface, such as the ad hit interface 210 of the system 200, and may be received by a component on the server side, such as the web server 214. After the advertiser submits the bid, process 600 may end.

Steps 606-610 may collectively indicate a bidding process 612 performed by an advertiser for an opportunity to display an advertisement to users of a social network. The bidding process 612 may be performed periodically by an advertiser (eg, monthly) for a particular advertisement. In some examples, the bidding process 612 may be automated for the advertiser based on the advertiser's previous bids in a similar advertisement. For example, if the advertiser provided average maximum and minimum bids for users in the first and fifth ranks of $ 1.00-0.50 and $ 0.25- $ 0.05, respectively, the bidding process 612 is part of the new bid, These average values may be automatically used for the first and fifth rank users.

In some examples, the automatic bid amount may be based at least in part on the user's influence score and ranking in the user's ranking list. For example, the automatic bid amount may be based on a bid amount per user influence score for the user rank (eg, bid amount / (user influence score / user rank)). Since user influence scores and user rankings may be provided to the advertiser in step 604, bid amounts may be automatically calculated for each user on the ranking list. In some examples, the threshold of the calculated bid amount (eg, bid amount greater than $ 0.01), user influence score (eg, score greater than average influence score), and / or user ranking (eg, rank in the top 25) may be determined by the user. Needs to be met in order to conduct automatic bidding.

7 is a flow chart of an example process for propagating an advertisement from a propagating user to a receiving user. Process 700 may be implemented by a system such as, for example, systems 100 and 200, and for clarity of representation, the following description uses system 100 and 200 as an example to describe processes. However, other systems or combinations of systems can also be used to implement the process 700.

Process 700 may begin at step 702 in which a user of a social network logs in to the social network. In step 704, the user can see the advertisement in the gadget. The gadget may be an interface of interaction included in the user's view of the social network, such as the gadget 212 described with reference to the system 200. For example, the gadget may be an interactive widget displayed in a user webpage based view of a social network. The gadget may display an advertisement that the advertiser has won the opportunity to display to the user. The gadget may also display advertisements propagated to the user from other users on the social network.

At step 706, process 700 may determine whether the user interacts with the advertisement being displayed in the gadget. User interaction may include clicking on an advertisement, providing an indication of interest in the advertisement (eg, pressing a “play” button for a video advertisement), and / or presenting the advertisement to other users of a social network. And if the user interacts with the advertisement displayed in the gadget, then process 700 proceeds to step 708. If the user does not interact with the advertisement displayed in the gadget, Process 700 does not propagate the advertisement at step 718.

In some implementations, step 706 further determines whether the process 700 ensures that the user's interaction propagates an advertisement to other users of the social network (eg, to users connected to the user by acquaintances). You can judge. For example, in the propagation of advertisements to occur, the user may indicate that he or she agrees to propagation through some kind of interaction with the gadget (eg, by clicking on a “propagation advertisement” button). If the user indicates consent for propagation, then process 700 can proceed to step 708. Although the user can interact with the advertisement, if the user does not support the propagation of the advertisement, the process 700 does not propagate the advertisement at step 718.

In step 708, process 700 determines whether the advertising budget has been exhausted. As described with reference to process 600, an advertiser may provide a total budget for displaying an advertisement. If the budget is exhausted by displaying the advertisement to users of the social network, the advertisement cannot be propagated. If the budget has not been exhausted, process 700 proceeds to step 710. If the budget is exhausted, process 700 does not propagate the advertisement at step 718.

In step 710, process 700 determines whether the time period for displaying the advertisement has expired. The advertiser may specify that the advertisement be displayed within a certain time period. For example, regarding an advertisement for a music concert, the advertiser may specify that the advertisement should be displayed by the concert time, not after the concert. If the time period for displaying the advertisement has not expired, process 700 proceeds to step 712. If the time period for displaying the advertisement has expired, process 700 does not propagate the advertisement at step 718.

In step 714, process 700 determines whether the advertisement has already been propagated to all users connected to the user by acquaintance. As described above, acquaintances, like friendships, post comments on other users 'profile pages, view other users' profile pages, social networking support activities (e.g. playing online games together via social networks, chat rooms, Interacting with other users through a conversation thread) and / or belonging to a similar social group or entity in the social network, and a connection between two users of the social network. If the advertisement has not been propagated to at least one of the users associated with the user by an acquaintance, process 700 proceeds to step 714. If the advertisement has already been propagated to all users connected to the user by acquaintance, then process 700 does not propagate the advertisement at step 718.

In step 714, process 700 determines whether a heat spread balance has been made between the user and the users associated with the user, to whom the advertisement has not yet been propagated and also linked by an acquaintance. As described above with reference to step 408 of process 400, an advertisement may be propagated from one user to another unless hit spread balancing is achieved. If the heat spread balance is not made in at least one user connected to the user by an acquaintance relationship, process 700 proceeds to step 716. If the heat spread balance is achieved for all users connected to the user by an acquaintance, process 700 does not propagate the advertisement at step 718.

In step 716, process 700 propagates an advertisement from a user (eg, a propagating user) to a receiving user (eg, a user associated with that propagating user by acquaintance relationship). Receiving users may be users who are acquainted with the propagating user, have an influence score less than the propagating user's influence score, and are users whose advertisement has not yet propagated. In some implementations, the advertisement can be propagated to all receiving users. In another implementation, the advertisement may be propagated only to receiving users who meet the evaluation criteria set by the advertiser. For example, the advertiser may instruct the propagation to the receiving user to occur only when the receiving user has an influence score greater than 80% of the propagating user's influence score. After propagating an advertisement for the receiving user, process 700 may end.

In step 718, process 700 does not propagate the advertisement to users connected to the user by acquaintance from the user. After step 718, process 700 proceeds to step 720. In step 720, process 700 determines whether there is a replacement advertisement that can be displayed to the user in the gadget. The alternate advertisement may be another advertisement that may be displayed to the user (eg, the advertiser wins the opportunity to display the alternate advertisement to the user, and the advertisement is propagated to the user). If the alternate advertisement can be displayed to the user, process 700 can return to step 704 and display the alternate advertisement to the user in the gadget. If a fallback advertisement is not available, process 700 may end.

System 800 includes a processor 810, a memory 820, a storage device 830, and an input / output device 840. Each of the components 810, 820, 830, and 840 are interconnected using a system bus 850. The processor 810 may process an instruction for execution in the system 800. The processor may be designed using some of a number of architectures. For example, the processor 810 may be a complex instruction set computer (CISC) processor, a reduced instruction set computer (RISC) processor, or a minimal instruction set computer (MISC) processor.

In one implementation, processor 810 is a single-threaded processor. In another implementation, the processor 810 is a multi-threaded processor. The processor 810 is capable of processing instructions stored in the memory 820 or on the storage device 830 to display graphical information for the user interface on the input / output device 840.

The memory 820 stores information in the system 800. In one implementation, memory 820 is a computer-readable medium. In one implementation, the memory 820 is a volatile memory unit. In another implementation, the memory 820 is a non-volatile memory unit.

Storage device 830 may provide a large amount of storage for the system 800. In one implementation, storage device 830 is a computer-readable medium. In various other implementations, storage device 830 can be a floppy disk device, hard disk device, optical disk device, or tape device.

Input / output device 840 provides input / output operations for the system 800. In one implementation, the input / output device 840 includes a keyboard and / or pointing device. In another implementation, the input / output device 840 includes a display unit for displaying a Graphic User Interface (GUI).

The features described above may be implemented in digital electronic circuitry, computer hardware, firmware, software, or in combinations of one or more of these. The apparatus may be embodied as a computer program product tangibly embodied for execution by an information transmitter, ie a machine-readable storage device or a programmable processor in a propagated signal; The method steps may be performed by a programmable processor executing a program of instructions to perform the functions of the implementations described above by operating on input data and generating output. The above described features in a programmable system comprising at least one programmable processor, at least one input device, and at least one output device coupled to receive data and instructions from a data storage system, and to transmit data and instructions. It may be advantageously implemented in one or more computer programs executable. A computer program is a collection of instructions that can be used directly or indirectly on a computer to perform any operation or to produce any result. A computer program may be used as a form of programming language that includes compiled or interpreted languages and may be included as a stand-alone program or module, component, subroutine, or other unit suitable for use in a computing environment. Can be deployed.

Processors suitable for the execution of the instruction program may include both general and special purpose microprocessors as a means of illustration, and either a single processor or a multiprocessor of a kind of computer. Generally, a processor will receive instructions and data from read-only memory or random access memory or both. The essential factor of the computer is a processor for executing instructions and one or more memories for storing instructions and data. In general, a computer will also include one or more mass storage devices for storing data files or be effectively coupled to communicate with the mass storage devices. Such devices include magnetic disks such as internal hard disks and removable disks; Magneto-optical disks; And optical disks. Storage devices suitable for tangibly embodying computer program instructions and data include, by way of example, EPROM, EEPROM, flash memory devices; Magnetic disks such as internal hard disks and removable disks; Magneto-optical disks; And all types of non-volatile memory, including semiconductor memory devices such as CD-ROM and DVD-ROM disks. The processor and memory may be complemented or integrated by application-specific integrated circuits (ASICs).

To provide for interaction with a user, the feature may be used to provide a display device, such as a cathode ray tube (CRT) or liquid crystal display (LCD) monitor, to display information to the user, and the user to provide input to the computer. Can be implemented on a computer with a keyboard and mouse or pointing device such as a trackball.

These features may include back-end components such as data servers, middleware components such as application servers or Internet servers, and front-end components such as client computers with graphical user interfaces or Internet browsers. front-end component) or a combination thereof. The components of the system may be connected by any form or medium of digital data communication such as a communication network. Examples of communication networks include local area networks (LANs), wide area networks (WANs), computers and networks that build the Internet.

The computer system may include a client and a server. Clients and servers are usually remote from each other and generally interact over the network as described. The relationship of client and server occurs by a computer program running on each computer and having a client-server relationship to each other.

Although some implementations have been described in detail above, other variations are possible. Other mechanisms for providing advertising to users of social networks may also be used. In addition, the logic flow shown in the figures does not necessarily require the particular order shown, or sequential order, to achieve desirable results. Other steps may be provided, or steps may be removed from the described flow, and other components may be added to or removed from the described systems. Accordingly, other implementations are within the scope of the following claims.

Claims (26)

As a computer-implemented method,
Receiving, at a server, information indicative of activity levels of computer-implemented social network users or acquaintances of the users on the computer-implemented social network;
Generating, by the server, influence scores for the users based on the received information; And
When the difference between the propagating user's first influence score and the receiving user's second influence score is greater than a threshold, by sending an advertisement from the propagating user to the receiving user And cyclically propagating the advertisement by a server through the computer-implemented social network among users with friends. The method according to claim 1,
After the advertisement is sent to the receiving user, the receiving user acting as a second propagating user. The method according to claim 1,
When the influence score of a particular propagating user is less than or equal to the influence score of a particular receiving user, terminating by the server the circular propagation of the advertisement through the computer-implemented social network. The method according to claim 1,
When the server receives an indication that a particular propagating user has refused the opportunity to propagate the advertisement via the social network implemented by the computer, the server terminates the cyclical propagation of the advertisement through the social network by the server. The method further comprises the step. The method of claim 4,
And the indication is generated by a client associated with the particular propagating user in response to the act of expressing indifference to the advertisement. The method of claim 4,
After the user refuses to interact, replacing the advertisement with a replacement advertisement by the server, and by the server, circularly replacing the replacement advertisement through the social network among users with whom they have a close relationship. Further comprising propagating. The method according to claim 1,
If the time period for the advertisement has expired, terminating, by the server, the cyclical propagation of the advertisement over the social network. The method according to claim 1,
When each of the receiving users having an acquaintance with a particular propagating user has an influence score greater than or equal to the influence score of the propagating user, the server terminates the cyclical propagation of the advertisement through the social network. The method further comprises a step. The method according to claim 1,
When each of the receiving users having an acquaintance with a particular propagating user has already received the transmission of the advertisement, terminating the cyclical propagation of the advertisement over the social network by the server. The method according to claim 1,
When the target number of propagation for the advertisement has been met, terminating, by the server, the cyclical propagation of the advertisement over the social network. The method according to claim 1,
When the budget for the advertisement is exhausted, terminating, by the server, the cyclical propagation of the advertisement through the social network. The method according to claim 1,
Providing, by the server, the propagating user an opportunity to interact with the advertisement. The method of claim 12,
When the propagating user chooses to propagate the advertisement by interacting with the advertisement, providing compensation by the server to the propagating user. The method according to claim 13,
Calculating, by the server, an amount of compensation based on the number of receiving users to which the advertisement has been propagated. The method of claim 12,
In a gadget disposed on a display of the propagating user of the computer-implemented social network, the opportunity to interact is provided to the propagating user. The method of claim 12,
The propagating user interacts with the advertisement by adding a comment about the advertisement or by explicitly specifying one or more users who will receive the advertisement based on interaction with the advertisement. The method according to claim 1,
Obtaining, by the server, data relating to the propagation of an advertisement through the computer-implemented social network from the propagating user and each receiving user. 18. The method of claim 17,
The propagation data includes information about the number of users who have viewed the advertisement, the number of users who have interacted with the advertisement, or the number of users who have propagated the advertisement. The method according to claim 1,
Recursively propagating the advertisement through the computer-implemented social network may include applying a heat diffusion model to the first impact score and the second influence score to determine the threshold value. And wherein the hit spread model is related to a path through which the advertisement can propagate through the computer-implemented social network. A system for distributing advertisements to users of a social network,
One or more servers;
An interface to the one or more servers for receiving information indicative of activity levels of social network users or acquaintances of users on the social network;
Means for generating influence scores for the user based on the received information; And
When the difference between the first influence score of the propagating user and the second influence score of the receiving user is greater than a threshold value, the social network is established among users with acquaintances by sending an advertisement from the propagating user to the receiving user. And a cyclical propagation module installed in the one or more servers to cyclically propagate the advertisement through. The method of claim 20,
The recursive propagation module allows the receiving user to act as a second propagating user after the advertisement is sent to the receiving user. The method of claim 20,
The propagation module terminates the cyclical propagation of the advertisement through the social network. As a computer-implemented method,
Outputting an advertisement to a first user of a computer-implemented social network, the output causing the advertisement to cause the advertisement to be displayed to the first user;
Receiving input related to the advertisement from the first user;
Retrieving influence scores for the first users, and influence scores for second users of the social network connected with the first user by acquaintance relations; And
Propagating the advertisement to the second users having an impact score less than the influence score for the first user. The method according to claim 23,
Generating influence scores for the first user, and influence scores for second users of the computer-implemented social network connected with the first user by an acquaintance relationship. The method of claim 24,
Generating an impact score for the user may be based on the user's acquaintances and the user's activity level on the computer-implemented social network for the other users of the computer-implemented social network. Calculating the amount of influence it has. The method according to claim 23,
The input from the first user includes an indication that the first user wants the advertisement to be propagated to the second users of the social network.

Images