Abstract
The purpose of word sense disambiguation (WSD) is to find the meaning of the word in any context with the help of a computer, to find the proper meaning of a lexeme in the available context in the problem area and the relationship between lexicons. This is done using natural language processing (NLP) techniques which involve queries from machine translation (MT), NLP specific documents or output text. MT automatically translates text from one natural language into another. Several application areas for WSD involve information retrieval (IR), lexicography, MT, text processing, speech processing etc. Using this knowledge-based technique, we are investigating Hindi WSD in this article. It involves incorporating word knowledge from external knowledge resources to remove the equivocalness of words. In this experiment, we tried to develop a WSD tool by considering a knowledge-based approach with WordNet of Hindi. The tool uses the knowledge-based LESK algorithm for WSD for Hindi. Our proposed system gives an accuracy of about 71.4%.
Highlights
Ambiguity is one of the characteristics of a natural language
Such strategies for word sense disambiguation (WSD) are generally connected to all words in a continuous text, in-spite of corpus-based methods, which are available for those words only for which annotated corpora are accessible
WSD finds its applications in many areas such as information retrieval (IR), information extraction (IE), and speech recognition (SR)
Summary
Ambiguity is one of the characteristics of a natural language. A word can have multiple meanings, varying with the context in which it is used. The WSD system can be interpreted to work as follows: it takes as input an arrangement of words or a sentence, NLP techniques are applied which utilize at least one source of knowledge to identify the most appropriate senses of the words regarding the context. Words possess different meanings that vary with the context in which they are used and our undertaking is to figure out which sense of word is intended in a particular context. This is one of the basic issues which are often experienced by any NLP framework. We revolve around approaches like selectional restriction with Hindi WordNet for WSD in case of Hindi language
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
