Semi-supervised Learning for Word Sense Disambiguation

TL;DR

This paper investigates various factors affecting unsupervised word sense disambiguation algorithms, proposes a lightly supervised version, and demonstrates comparable performance to optimized methods through a pseudo-word evaluation strategy.

Contribution

It identifies key factors influencing the algorithm's performance and introduces a lightly supervised approach with an evaluation strategy for improved disambiguation.

Findings

Factors like initial labeling and rule confidence significantly impact performance

The lightly supervised method achieves results comparable to optimized algorithms

Pseudo-word evaluation effectively measures the impact of different factors

Abstract

This work is a study of the impact of multiple aspects in a classic unsupervised word sense disambiguation algorithm. We identify relevant factors in a decision rule algorithm, including the initial labeling of examples, the formalization of the rule confidence, and the criteria for accepting a decision rule. Some of these factors are only implicitly considered in the original literature. We then propose a lightly supervised version of the algorithm, and employ a pseudo-word-based strategy to evaluate the impact of these factors. The obtained performances are comparable with those of highly optimized formulations of the word sense disambiguation method.