Inferring professional reputations of social network members

A system comprising:

• one or more computer processors;
• one or more modules incorporated into an online social networking service to improve a capability of the online networking service to identify a content item stored in a database of the online social networking service as a spam content item, the one or more modules configuring the one or more computer processors to perform operations, the operations comprising:
• generating a recommendation graph represented as a matrix listing vertices on both the vertical and horizontal axes and edges as entries corresponding to an intersection of the vertices, each of the vertices representing respective electronic profiles of members of the social networking service, each of the edges representing a recommendation of a recommendee member of the social networking service by a recommender member of the social networking service;
• training a reputation model to learn a respective importance for each respective feature of a subset of features of the electronic profiles, the training including providing the generated recommendation graph to a classifier;
• estimating a professional reputation of a member, the estimating including applying the trained reputation model to a feature vector of the member, the feature vector including feature values for the subset of the features, the applying including adjusting at least one of the feature values by a respective weight corresponding to the respective learned importance of the respective feature;
• aggregating a set of estimated professional reputations of members that have engaged with the content item stored in the database of the online social networking service;
• performing the identifying based on the aggregated set of estimated professional reputations;
• and
• determining a target audience for the content item based on the identifying.

Anon-transitory machine-readable storage medium including instructions, which, when executed by one or more computer processor of a machine, cause the machine to perform operations, the operations comprising:

• generating a recommendation graph represented as a matrix listing vertices on both the vertical and horizontal axes and edges as entries corresponding to an intersection of the vertices, each of the vertices representing respective electronic profiles of members of the social networking service, each of the edges representing a recommendation of a recommendee member of the social networking service by a recommender member of the social networking service;
• training a reputation model to learn a respective importance for each respective feature of a subset of features of the electronic profiles, the training including providing the generated recommendation graph to a classifier;
• estimating a professional reputation of a member, the estimating including applying the trained reputation model to a feature vector of the member, the feature vector including feature values for the subset of the features, the applying including adjusting at least one of the feature values by a respective weight corresponding to the respective learned importance of the respective feature;
• aggregating a set of estimated professional reputations of members that have engaged with the content item stored in the database of the online social networking service;
• performing the identifying based on the aggregated set of estimated professional reputations;
• and
• determining a target audience for the content item based on the identifying.