Computing a ranked feature list for content distribution in a first categorization stage and second ranking stage via machine learning
Abstract:
An online system identifies seed users with a high value score to a third party system. The online system identifies features of each of the seed users. A weight for each of the identified features is identified. The identified features are divided into a plurality of buckets, each bucket indicating a property associated with one or more of the identified features. Each bucket is ranked according to the weights of the identified features in each bucket. The online system identifies an additional user that has a threshold measure of similarity the seed users. The online system transmits a content item to the additional user for presentation. Additionally, the online system transmits one or more third party-presentable factors based on the bucket having the highest rank to the third party system, the third party-presentable factors indicating a reason as to why the additional user was presented with the content item.
Information query
Patent Agency Ranking
0/0