Graph based Word Sense Disambiguation for Gujarati

Abstract

Due to high Semantic Ambiguity that is associated with language, Word Sense Disambiguation is an open problem of Natural Language Processing and Ontology in Computer Linguistics. For instance, the Gujarati word પુવૅ (Purva) can mean East and Past. The task of automatically assigning the most suitable meaning to a polysemous word can be defined as Word Sense Disambiguation. Several approaches have been done, from Dictionary-and Knowledge-based method that uses Lexical resource like WordNet, to Supervised-Machine learning method that use trained classifier on manually sense-annotated corpus, to cluster based unsupervised methods. In this approaches knowledge based approach  is suffer from knowledge acquisition bottleneck problem, While Supervised approach requires large amount of manually sense tagged corpora which are highly expensive and need human resource. These approaches have been applied for several languages like English, French, German, Chinese, and some Indian languages like Hindi, Marathi, Assamese, Malayalam, however research on Word Sense Disambiguation for Gujarati language is relatively limited. There for we have proposed Unsupervised Graph based Word Sense Disambiguation for Gujarati where Indo WordNet for Guajarati is used as lexical database for WSD.

Keywords: Word sense disambiguation, natural language processing, ambiguity, WordNet

Cite this Article
Manali Parekh, Prem Balani, Deven Gol,Graph Based Word Sense Disambiguation for Gujarati. Journal of Advancements inRobotics. 2018; 5(2): 37–45p.