Heuristic Computational Matrix Method for Marathi Grammar Checker
Nivedita S. Bhirud1, R.P.Bhavsar2, B.V.Pawar3
1Nivedita S. Bhirud*, Department of Computer Engineering, Vishwakarma Institute of Information Technology, Pune, India.
2R.P. Bhavsar, School of Computer Sciences, Kavayitri Bahinabai Chaudhari North Maharashtra University, Jalgaon, India.
3B.V. Pawar, School of Computer Sciences, Kavayitri Bahinabai Chaudhari North Maharashtra University, Jalgaon, India.
Manuscript received on December 12, 2019. | Revised Manuscript received on December 21, 2019. | Manuscript published on January 10, 2020. | PP: 1540-145 | Volume-9 Issue-3, January 2020. | Retrieval Number: C8581019320/2020©BEIESP | DOI: 10.35940/ijitee.C8581.019320
Open Access | Ethics and Policies | Cite | Mendeley
© 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: Spelling, morphology, syntax and semantics are the important areas of Natural Language (NL) sentence analysis. Syntax checking of a sentence is broadly referred as a ‘grammar checking’, however it also involves morphological analysis hence technically it is a multidimensional problem. Syntax of a natural language defines permissible sentence structures and constraints on constituents such as their order and unification constraints. It is a purely theoretical aspect and considered as computationally trivial rule enforcement problem. Rule formulation needs expert labour work and is costly and time consuming affair. Modern data driven language engineering approach advocates use of minimal knowledge base (linguistic information) and relies on knowledge extraction from tagged data. It is difficult to find such tagged data for non-English natural languages like Marathi (Indian Language). Considering these facts for grammar checking problem, we have come up with intuitional heuristic method for Marathi grammar checking which uses basic syntactic cues and minimal lexical information. We have modeled this heuristic method scientifically using basic matrix comparison operation. Our approach relies on syntactic cues like word ending, verb ending. We have tested our method on handcrafted Marathi sentences catering different Marathi sentence structures (one hundred and fifty three). The performance is measured using precision and recall metrics. The system has yielded 83% precision and 93% recall on sample data. This approach can be exploited for well structured text documents typically in the closed domains like legal, official, educational etc.
Keywords: Computational Linguistics, Heuristic Function, Marathi Language Grammar, Natural Language Processing, Rule based approach, Statistical approach
Scope of the Article: Natural Language Processing