Myanmar Noun Phrase Identification System for Myanmar to English Translation

Abstract

Noun Phrase Identification is one of the most critical components of natural language processing because information retrieval systems of all kinds rely on base-noun phrase as their primary source of entity identification. Translating noun phrase is also a very important task toward sentence translation since noun phrases form the majority of textual content of the document. In this paper, we implement a Myanmar noun phrase identification algorithm which uses the rule based approach. The system uses the linguistic theory of Myanmar language for noun phrase which is attached with prepositional phrase, modified with adjective and adverb. To identify the Myanmar noun phrase from a sentence, Context Free Grammar rules, which are generated by using the part of speech sequences, are used. The system accepts the Myanmar tagged sentence as an input. After identifying the noun phrases in the sentence, Myanmar noun phrases are generated as the output. The algorithm is also evaluated using tagged sentences.