Author(s)

S. Massa & P.P. Puliafito

Abstract

Data Mining is a new research area that develops techniques for knowledge discovery in massive amounts of data. The way such data are collected and modeled is quite important to define which kind of knowledge could be investigated and consequently to develop proper methods. One of the most investigated topics in data mining is the discovery of knowledge about data organized as temporal sequences. This work proposes a new modeling technique to mine information from data that are expressed in the form of temporal sequences of events associated to entities. In particular such a technique aims at extracting non-evident behavioral patterns from data in order to identify different classes of entities in the considered population. In order to represent the behavior of the entities, a Markov cha

Keywords