Word Sense Disambiguation for Gujarati Language

Abstract

newline Word Sense Disambiguation (WSD) is the process of identifying which sense of a word is used in a given sentence. A word can have several senses, which is termed as an ambiguity. Something is ambiguous when it can be understood in two or more possible ways or when it has more than one meaning. This word sense disambiguation is an intermediate task , which is not an end in itself, but rather is necessary at one level or another to accomplish most natural language processing tasks. newline newlineWord Sense Disambiguation (WSD) is an important concept that is to be evaluated for performing machine translation, and a tool is needed to perform disambiguation so that computers would be able to interpret a word in its proper sense according to its context. While the work on automatic WSD for English is voluminous, this is the first attempt for Gujarati WSD. newline newlineIn this thesis, the proposed method 1 is used for disambiguating word sense based on the context using the Gujarati language. A word sense disambiguation algorithm is implemented for Gujarati words with the help of the Gujarati wordnet. The proposed method 2 using Gujarati wordnet and Gujarati web resources like Gujarati wiktionary are used to train the Gujarati language model and estimate similarity between Gujarati sentences context-wise. newline newlineThe main idea is to compare the context of the word in a sentence with the contexts constructed from the WordNet and chooses the winner. The mentioned WordNet contexts are built from semantic relations and glosses. The results of experiments show that the accuracy obtained by the implementation is 73% and can be improvised by adding some extra computations and training data to the algorithm for the Gujarati language. newline newline newline