Second International Conference on Spoken Language Processing (ICSLP'92)

Banff, Alberta, Canada
October 13-16, 1992

Learning Compatibility Coefficients for Word-Class Disambiguation Relaxation Processes

Marcello Pelillo (1), Mario Refice (2)

(1) Department of Computer Science - University of Bari, Bari, Italy
(2) Department of Electrical and Electronic Engineering - Polytechnic of Bari, Bari, Italy

Relaxation labeling processes are iterative procedures that aim at reducing local ambiguities through a parallel exploitation of contextual information. Recently they have been successfully applied to the problem of word-class disambiguation. In relaxation processes contextual information is embedded in a set of "compatibility coefficients." Several statistical-based interpretations have been given for compatibility coefficients, and recently a new optimization standpoint has been introduced. In this paper we discuss these two approaches for learning compatibilities on the basis of a training sample, and suggest a way to combine them. Experimental results conducted over a sample text are given.

Full Paper

Bibliographic reference.  Pelillo, Marcello / Refice, Mario (1992): "Learning compatibility coefficients for word-class disambiguation relaxation processes", In ICSLP-1992, 389-392.