K­MEANS CLUSTERING FOR HIDDEN MARKOV MODELS

Michael P. PERRONE and Scott D. CONNELL

Pen Technologies Group,
IBM T.J. Watson Research Center
mpp@us.ibm.com

An unsupervised k­means clustering algorithm for hidden Markov models is described and applied to the task of generating subclass models for individual handwritten character classes. The algorithm is compared to a related clustering method and shown to give a relative change in the error rate of as much as 8% on a 30,000­word vocabulary, unconstrained­ style, on­line, writer­independent handwriting recognition task.

In: L.R.B. Schomaker and L.G. Vuurpijl (Eds.)
Proceedings of the Seventh International Workshop on Frontiers
in Handwriting Recognition, September 11-13 2000, Amsterdam,
Nijmegen: International Unipen Foundation,
ISBN 90-76942-01-3
pp. 229-238.