WORD SENSE DISAMBIGUATION : AN INTEGRATED FRAMEWORK

Abstract

Word Sense Disambiguation or WSD intent to find the exact meaning of an ambiguous word for a particular context. Its enormous applications lies in various research areas including sentiment analysis, Information Retrieval, Machine translation and knowledge graph construction. The main objective remains intact that words with same spelling can have completely different senses depending on subtle characteristics of the context. In this paper we analyse the traditional word expert supervised methods. However in comparison with knowledge-based methods, supervised method outperforms better. We standardise the pre-trained BERT and LESK algorithms on SemEval data set and experiment the algorithms on this dataset & compare the accuracy for better results.