Automatic construction of personalized, peer-derived messages for mobile health applications

摘要

A method is implemented by at least one server. The method includes obtaining a set of users as a function of profile records regarding a plurality of users. The method includes defining a subset of the users as a peer group based on peer group formation features by applying a clustering algorithm to profile records of the set of users. The method includes calculating a peer group metric value for the peer group based on behaviors of the subset of users in the peer group. The method includes calculating a user metric value for a target-user included in the subset based on behavior of the target-user. The method includes constructing a message relating the patient metric value to the peer group metric value.