Recognition and disambiguation of polysemy words in entered hindi documents

摘要

In this digital age, natural language is causing hindrance in the advancement of information technology revolution in India. There is a need to perform Natural Language Processing (NLP) using computer processing, so that computer based systems can be accessible through natural languages like Hindi. Therefore a language translator is very important tool to resolve this problem. One of the key challenges involved in the design of a language translator is Polysemy disambiguation. Polysemy means having two or more meanings to a single word. For language translation operations, it is crucial to find the right meaning of any given word in its context. This is known as Word Sense Disambiguation (WSD). Various resources and tools already exist for WSD in English language while regional languages have not been equally explored yet. This project has a user interface where the user enters a meaningful sentence in Hindi with polysemy word. The system identifies the polysemous word/s if any in the sentence and lists one or more meanings associated with the polysemous word. Using existing machine learning techniques the project identifies the right meaning of the polysemous word based on the given context.