Learning Process Behavior with EDY: an Experimental Analysis
Ugo Galassi · Frontiers in artificial intelligence and applications · 2008
This paper presents an extensive evaluation, on artificial datasets, of EDY, an unsupervised algorithm for automatically synthesizing a Structured Hidden Markov Model (S-HMM) from a database of sequences. The goal of EDY is capturing the stochastic process by which the observed data was generated. The SHMM is a sub-class of Hidden Markov Model that exhibits a quasi-linear computational complexity and is well suited to real-time problems of process/user profiling. The datasets used for the evaluation are available on the webhttp://www.edygroup.di.unipmn.it