Abstract

In this paper, we discuss a framework for weighted combination of classifiers for word sense disambiguation (WSD). This framework is essentially based on Dempster–Shafer theory of evidence [G. Shafer, A Mathematical Theory of Evidence, Princeton University Press, Princeton, 1976] and ordered weighted averaging (OWA) operators [R.R. Yager, On ordered weighted averaging aggregation operators in multicriteria decision making, IEEE Transactions on Systems, Man, and Cybernetics 18 (1988) 183–190] We first determine various kinds of features which could provide complementarily linguistic information for the context, and then combine these sources of information based on Dempster’s rule of combination and OWA operators for identifying the meaning of a polysemous word. We experimentally design a set of individual classifiers, each of which corresponds to a distinct representation type of context considered in the WSD literature, and then the discussed combination strategies are tested and compared on English lexical samples of Senseval-2 and Senseval-3.

Keywords: Computational linguistics; Classifier combination; Word sense disambiguation; OWA operator; Evidential reasoning

Article Outline

1. Introduction

2.Preliminaries

2.1. Dempster–Shafer theory of evidence

2.2. OWA operators

3. Weighted combination of classifiers for WSD

3.1. WSD with multi-representation of context

3.2. DS theory based combination scheme

3.2.1. The discounting-and-orthogonal sum combination strategy

3.2.2. The discounting-and-averaging combination strategy

3.3. OWA operator based combination scheme

3.3.1. Max rule

3.3.2. Min rule

3.3.3. Median rule

3.3.4. Fuzzy majority voting rules

3.3.5. Majority vote rule

4. Experimental study

4.1. Representations of context for WSD

4.2. Test data

4.3. Experimental results

5. Conclusions

Acknowledgements

References

Vitae