WORD SENSE DISAMBIGUATION USING FUZZY SEMANTIC-BASED STRING SIMILARITY MODEL

Abstract

Sentences are the language of human communication. This communication medium is so fluid that words and meaning can have many interpretations by readers. Besides, a document that consists of thousands of sentences would be tough for the reader to understand the content. In this case, computer power is required to analyse the gigantic batch size of the text. However, there are several arguments that actively discuss regarding the output generated by a computer toward the meaning of the passage in terms of accuracy. One of the reasons for this issue is the existing of the ambiguous word with multiple meanings in a sentence. The passage might be incorrectly translated due to wrong sense selection during the early phase of sentence translation. Translating sentence in this paper means either the sentence has a negative or positive meaning. Thus, this research discusses on how to disambiguate the term in a sentence by referring to the Wordnet repository by proposing the use of fuzzy semantic-based similarity model. The proposed model promising to return a good result for detecting the similarity of two sentences that has been proven in the past research. At the end of this paper, preliminary result which shows the flow of how the proposed framework working is discussed.