Cumulative Mean Function of Public Posting Episodes in the Online Media with Regard to User’s Digital Traces: Limited Data on publications Dates and Profile Data

Abstract

The problem of behavior associated risk assessment arises in some areas that are closely related with a person’s behavior. The data for this problem is often incomplete, what brings the problem into the soft computing framework. In the paper the cumulative mean function is considered as the characteristic, reflecting the behavior pattern. Two approaches are presented for the modelling of this function in the context of gamma Poisson mathematical model of behavior: the nonparametric approach and regression-based one. Various observed and unobserved individual characteristics can be taken into account in the regression model. The approaches are illustrated with the data on public posting in the online media.