Skeletal Algorithms: Sequential Pattern Mining

Abstract

The basic idea behind the skeletal algorithm is to express a problem in terms of congruences on a structure, build an initial set of congruences, and improve it by taking limited unions/intersections, until a suitable condition is reached. Skeletal algorithms naturally arise in the context of data/process mining, where the skeleton is the “free” structure on initial data and congruence corresponds to similarities in data. In this paper we study skeletal algorithms applied to sequential pattern mining and compare their performance with real models, Markov chains and models based on Shannon entropy.