Facebook, Inc.

Yanxin Shi

David Chen

Florin Ratiu

Kiranjit Singh Sidhu

Emily B Grewal

Denise Moreno

Ching-Chih Weng

Lex Arquette

Huan Yang

Andrey Goder

David Dawei Ye

Yigal Dan Rubinstein

Kent Schoen

Srinivas P Narayanan

Levy Klots

Hong Yan

Robert Kang-Xing Jin

Gideon Wald

John Hegeman

Wenjie Fu

Alexey N Spiridonov

Robert Jin

Alexey Spiridinov

A social networking system presents advertisements and recommendation units to its users. The recommendation units suggest actions for the users to increase their engagement with the social networking system or otherwise interact with other users, while the social networking system receives revenue from advertisers for displaying advertisements based on bid values associated with the advertisements. The social networking system determines values for the advertisements and for the recommendation units, where the values are measured in a comparable fashion. This allows the system to rank and select the advertisements and recommendation units together in a unified auction model. For example, the social networking system uses a pacing value to determine values of recommendation units having a common unit of measurement with expected values of advertisements to the social networking system.

Unified auction model for suggesting recommendation units and ad units

Techniques to optimize messages sent to a user of a social networking system. In one embodiment, information about the user may be collected by the social networking system. The information may be applied to train a model for determining likelihood of a desired action by the user in response to candidate messages that may be provided for the user. The social networking system may provide to the user a message from the candidate messages with a selected likelihood of causing the desired action.

Methods and systems for optimizing messages to users of a social network

In one embodiment, a system includes one or more computing systems that implement a social networking environment containing a large number of heterogeneous objects type, each of the plurality of object types having varying features, the system implementing a generic object recommendation engine for scoring objects and recommending the objects to users of the social networking system. In particular embodiments, the user and content object features are fed as inputs into a heuristic model that generates an expected value for the content object and user. In particular embodiments, the object recommendation engine includes an online learner that may log a user's actions after the initial impression to determine the relatively degree of interest to the user.

Real-time online-learning object recommendation engine

Optimizing messages to users of a social network using a prediction model that determines likelihood of user performing desired activity

Optimizing messages to users of a social network using a message that includes a user's performance of a desired activity associated with a link included in the message

As a user of a social networking system views a page that includes information provided by the system, certain types of social interactions are monitored. If an interaction monitored for is detected, at least one recommendation unit is identified to present to user on the page. The recommendation unit is identified based on a description of the interaction. The recommendation unit suggests that the user perform a social interaction in the social networking system. The recommendation unit is transmitted to a device of the user and is presented to the user on the page without having to reload the entire page.

Instantaneous recommendation of social interactions in a social networking system

Methods, apparatuses and systems for generating a nickname dictionary that includes associations between names of users and candidate nicknames based on statistical analysis of user communications observed at a network communications facility, such as a social network system, an email provider and the like.

Generation of nickname dictionary based on analysis of user communications

Growing community of inventors

Palo Alto, CA, United States of America

Yanxin Shi