Detection of Synonyms from Myanmar Text Using Latent Semantic Analysis

Abstract

Manually produced lexical resources are time-consuming and labour-intensive. Therefore, automatically identifying these resources is important to overcome knowledge bottleneck in natural language processing. This paper deals with the task of finding synonyms and near synonyms from raw data for the lexical substitution task. In this paper, an approach for automatically identifying synonyms is presented to support the process of semantic simplification for Myanmar text. Candidate synonyms for a target word will be generated using latent semantic analysis (LSA), which is a fully automatic statistical technique for extracting relations of expected contextual usage of words. This work can be helpful for thesaurus construction for Myanmar language.