Open AccessImplementation of Automatic Mesh Rules Generation by Word Features Relationship (AMRG-WFR) for Word Sense Disambiguation for Marathi Language
Author(s) -
Swati Kale,
Ujwalla Gawande
Publication year - 2021
Publication title -
international journal of next-generation computing
Language(s) - English
Resource type - Journals
eISSN - 2229-4678
pISSN - 0976-5034
DOI - 10.47164/ijngc.v12i5.456
Subject(s) - computer science , natural language processing , artificial intelligence , wordnet , semeval , word (group theory) , automatic summarization , marathi , context (archaeology) , machine translation , natural language , feature (linguistics) , task (project management) , linguistics , paleontology , philosophy , management , economics , biology
One common feature of natural languages is a single word having the different meanings. Detecting the correct sense of the word in the given context is known as word sense disambiguation. In Natural language processing WSD is always a challenging and difficult task. The accuracy of many systems like information retrieval, machine translation text summarization depends upon the module used to identify the correct sense of the word. In this paper we proposed an automatic mesh rule generation technique by word feature relationship for the task of word sense disambiguation for Marathi language. Marathi is one the most popular and morphologically rich language in India. We conducted all the experiments on the benchmark WordNet dataset developed at Indian Institute of Technology Bombay (IIT Bombay). The proposed algorithm contains three sections.In the first section supervised learning technique is used to create training data set. Mesh rules of object were created in second section and in last section the correct sense of ambiguous word matched with the given context. Results show that the proposed method has superior performance as compared with already developed approaches.