Chunking Marathi Text using Marathi Grammar Rules and Conceptual Dependency Representation
Madhuri M. Deshpande1, Sharad D. Gore2
1Mrs. Madhuri M. Deshpande, Department of Computer Science, Savitribai Phule Pune University (formerly University of Pune), Pune, India.
2Dr. Sharad D. Gore, Department of Computer Science, Savitribai Phule Pune University (formerly University of Pune), Pune, India.
Manuscript received on 23 November 2016 | Revised Manuscript received on November 2016 | Manuscript published on 30 November 2016 | PP: 5-11 | Volume-5 Issue-5, November 2016 | Retrieval Number: E1631115516©BEIESP
Open Access | Ethics and Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: The paper aims at using a rule based chunker to create chunks, such as noun phrase (NP) and verb phrase (VP), for a given Marathi sentence. Chunking is the process that labels segments of a sentence with syntactic constituents such as noun/verb/adjective phrases (NP, VP, AP). Chunking is an important task in Natural Language Processing (NLP). We have used a modified YASS POS tagger to tag the input Marathi sentence. We have applied Marathi grammar rules, in the form of regular expressions, and used the Conceptual Dependency (CD) theory representation to represent the dependency of words in Marathi sentence, which ultimately would depict the meaning of words in the sentence with respect to the context in which a bag-of-words are used. A rule-based chunker create chunks in Marathi sentence. Conceptual Dependency Theory focuses on concepts and understanding about a concept instead of syntax and structure.
Keyword: Chunking, Conceptual Dependency, Dependency Parser, Natural Language Processing.
Scope of the Article: Natural Language Processing