Facebook, Inc.

Meta Platforms, Inc.

Spencer Powell

Zhengyong Zhu

Jiajin Yu

Patrick James Kemp

Weidong Wang

Roshan Umesh Goli

Max K Comer

Kevin Wenkai Heh

Boris Pierre Arnoux

Prathyusha Aragonda

Sumeet Kumar

A social networking system maintains characteristics with its users, with various characteristics, such as age, specified by the users (i.e., 'asserted characteristics'). The social networking system selects content for a user based at least in part on the characteristics associated with the user. To account for potential inaccuracies in an asserted age of a user, the social networking system clusters users based on ages of other users connected to users. The online system receives verified ages for users in a cluster from a trusted third party system that maintains more accurate characteristics for users than the social networking system. By comparing the asserted ages for users in the cluster to the verified ages for users in the cluster, the social networking system determines an accuracy of the asserted ages for users in the cluster. The accuracy may be used when selecting content for the users.

Determining accuracy of characteristics asserted to a social networking system by a user

An online system receives a video-presentation request from a third party system. The video-presentation request comprises a video and a target audience specification for the video. The online system selects a plurality of users as the target audience of the video based on the target audience specification. From the target audience, the online system generates a sample subset of users and determines a sampled video reach count for the subset of users. A sample user who, as determined by the online system, would have viewed the view for at least a threshold duration, is included in the sampled video view count. The online system estimates a total video reach count for target audience by extrapolating the sampled video reach count for the subset of users to the target audience. The total video reach count can be used to determine a parameter for presenting the video on an online system.

Computation of video presentation statistics in an online system

An online system predicts a performance of a content item campaign based on a set of received target parameters from a content item publisher and compares the predicted performance of the content item campaign to predicted performances of comparison campaigns. The comparison campaigns are selected by performing a plurality of simulations on prior content item campaigns to estimate campaign presentation results for each prior content item campaign if presented to users with the sets of test campaign parameters. After generating histograms based on the estimated results, the online system selects prior content item campaigns for the set of comparison campaigns based on changes in histogram positions of the prior content item campaigns. The online system generates a set of estimated campaign presentation results for the content item campaign and the set of comparison campaigns for comparison.

Predicting performance of content item campaigns

An online system scores campaign audiences based on historical scoring data for similar audiences. A third party system selects a target audience and a day on which the target audience should be exposed to a campaign. The online system generates an availability grid and a score grid to determine a score for the target audience. Values in the availability grid are determined based on the availability of exposure time for the target audience on the specified date. Values in the score grid are based on historical scoring data for the same audience. The online system scores the target audience by interpolating between data points in the score grid based on a selected availability from the availability grid.

Using interpolation based on historical data to generate a time-agnostic scoring for users in an online system

An analysis system analyzes known user affinities to identify particular objects that serve as useful predictors of whether a given user will have an affinity for a given topic, even if the user has not previously expressly specified an affinity for that topic. Specifically, both the topic group of users associated with a given topic and the category group of users associated with the topic's more general category are identified. For each of a set of objects, degrees of divergence between the topic group and the category group are evaluated for a criterion evaluated with respect to the object. A topic profile is created based on objects for which there is a high degree of divergence.

Formation of topic profiles for prediction of topic interest groups

A system forms sub-groups from a given user group of a social networking system and form descriptions of the sub-groups that provide an intuitive understanding of sub-group composition, such as likings of the sub-groups. In one embodiment, a given user group of a social networking system is clustered into a plurality of sub-groups, and representative characteristics&#x2014;such as the characteristics of a composite or actual member of the sub-group&#x2014;are determined for each sub-group. In order to form sub-group descriptions, a set of objects, such as pages of the social networking system, is ranked with respect to the representative characteristics of the sub-group. The highest-ranking objects for a sub-group are then used to form the description of that sub-group. For example, the topics associated with each of the highest-ranking pages can be combined into the sub-group description.

Growing community of inventors

San Francisco, CA, United States of America

Spencer Powell