Sentence Reduction for Syntactic Analysis of Compound Sentences in Punjabi Language

Abstract

Objective: This research paper is an attempt to develop a syntactic analysis system for compound sentences of Punjabi language. Methods/Statistical Analysis: Sentence simplification approach has been used for splitting compound sentences into simple sentences and then analyzing these simple sentences for syntactical error. A full form lexicon based morph, HMM based POS tagger and set of rules have been used for identification of grammatical mistakes. Findings: On testing an overall precision as 93.30, recall as 97.32 and F-measure as 95.25 is reported by the system. Application/Improvements: the system shows better performance on comparing it with existing Punjabi grammar (Precision=76.79, Recall=87.08) and Myanmar grammar checker that works on compound sentence and shows precision 83.75%.