Reward-based recommendations of actions using machine-learning on telemetry data

Abstract

System and methods for automatically providing action recommendations are described. A method may include collecting a set of telemetry data from a client application. The set of telemetry data contains a plurality of pages generated based on a plurality of user actions performed on the client application. The method may include generating a plurality of prior probabilities corresponding to the plurality of pages and the plurality of user actions. In response to the client application displaying a first page, the method may generating a plurality of posterior probabilities for a subset of user actions that can be invoked in the client application from the first page, and selecting a plurality of recommended actions from the subset of user actions for having the highest corresponding posterior probabilities among the plurality of posterior probabilities.