Systems and methods for optimizing advertising spending using a user influenced advertisement policy
Abstract:
Systems and methods are disclosed for generating an optimized bid value for an advertisement relating to a product. In one implementation, a computerized-method is provided for receiving information about online activity by a plurality of users, the information including a user ID associated with each of the plurality of users. The method may also include extracting a feature set corresponding to purchasing users, the purchasing users being a subset of the plurality of users who have purchased the product. The method may further include generating a model that represents the extracted feature sets of the purchasing users, based on the extracted data, calculating a reward function according to the generated model, and generating an optimized bid value for the advertisement according to the calculated reward function, the optimized bid value representing the difference between a value of displaying the advertisement, and a value of not displaying the advertisement.
Information query
Patent Agency Ranking
0/0