Word Sense Disambiguation (WSD) became a challenge the very day machine translation was started being attempted. The need for disambiguating competing senses of ambiguous words is a crucial issue for all the Natural Language Processing activities including machine translation. The source word with multiple senses has to be disambiguated before resorting to lexical transfer from source language to target language. It has to be done by default that Tamil words have to be disambiguated before translating the Tamil text into English or any other languages. Disambiguating word senses found in texts, from the computational point of view, is a classificatory process of discriminating one sense from the other. As the sense interpretation rely on the context, the classification of contexts based on the senses becomes crucial. Support Vector Machine (SVM) comes handy for this effort. The SVM will do the classificatory process of discriminating the contexts there by selecting the correct sense of the target word. In this supervised frame-work, a small number of annotated examples for each sense of the target word are used for training the SVM classifier. The system is found to be efficient if training is done with efficiently annotated text and good feature selection. © Research India Publications.
cited By 2
A. M. Kumar, Rajendran, S., and Soman, K. P., “Tamil word sense disambiguation using support vector machines with rich features”, International Journal of Applied Engineering Research, vol. 9, pp. 7609-7620, 2014.