WORD SENSE DISAMBIGUATION IN THE HINDI LANGUAGE: NEURAL NETWORK APPROACH

Abstract

Hindi is the national language of India. A massive number of peoples share, retrieve, and access documents in the Hindi language. Hindi Word Sense Disambiguation (HWSD) system used to extract ambiguity from the Hindi language. “Word Sense Disambiguation (WSD) eliminates ambiguity and you can easily understand the meaning of a specific ambiguous word used in sentence”. It comes up as a field of research in computational linguistics and it helps in learning the real concept of the words appearing in a particular context. Humans can easily use the WSD technique to distinguish the different meanings and can speak a better language. However, computers may find it difficult to deal with the WSD technique. There are different approaches using which it has become easy to carry out the complete procedure. The four main approaches, which are commonly used, are knowledge-based, Supervised, Semi-Supervised, and Unsupervised. Hence, it improves the computer’s performance and you can learn the true importance of search engine optimization. It also helps in collecting information and helps in dealing with different software’s. If you are looking for a voice assistant this method works the best and you can explore the best form of machine learning. It comes up with an organized neural network and the algorithms help in detecting the differences easily and you would get accurate results. There is an inner layer of the network with nodes and you can recognize the binary values, which are set according to the frequency of the context words followed by the ambiguous words. On the other hand, there is an outer layer too consisting of the nodes, which has a similarity to the senses of the ambiguous words. “In this paper, we describe different approaches used in WSD, resources required for disambiguation tasks, and a review of previous works for the Hindi language”.