next up previous index
Next: Comparison of the probability Up: Variable length Markov model Previous: Variable length Markov model   Index





Introduction

The aim of this chapter is to present the results we obtained in the evaluation of the variable length Markov model. The first section compares the different way of estimating probabilities. The second section compares the use of Kullback-Leibler (KL) and Matusita distances in the VLMM algorithm. And finally an assessment of the prediction capabilities of a learned VLMM tree is given.



franck 2006-10-16