One Sense per Translation

TL;DR

This paper introduces the One Sense per Translation principle, establishing theoretical properties for translation-based sense inventories and demonstrating a WSD method with high precision and improved accuracy on challenging datasets.

Contribution

It defines necessary conditions for translation-based sense inventories and proposes a WSD method grounded in the One Sense per Translation property.

Findings

Achieves approximately 93% precision in intrinsic evaluation.

Demonstrates up to 4.6% F1-score improvement on difficult WSD datasets.

Establishes theoretical properties for translation-based sense delimitation.

Abstract

Word sense disambiguation (WSD) is the task of determining the sense of a word in context. Translations have been used in WSD as a source of knowledge, and even as a means of delimiting word senses. In this paper, we define three theoretical properties of the relationship between senses and translations, and argue that they constitute necessary conditions for using translations as sense inventories. The key property of One Sense per Translation (OSPT) provides a foundation for a translation-based WSD method. The results of an intrinsic evaluation experiment indicate that our method achieves a precision of approximately 93% compared to manual corpus annotations. Our extrinsic evaluation experiments demonstrate WSD improvements of up to 4.6% F1-score on difficult WSD datasets.