Detecting and Correcting Contextual Mistakes in Sentences Using Part of Speech Tags

Abstract

A grammar checker is a tool to check each sentence in a text to see whether it conforms to the grammar. In case it finds a structure that conflicts with the conformity to the grammar, it would give suggestions for alternatives. The grammar checkers for European languages and some Indic languages are well developed. However, perhaps, owing to Tamil being a morphologically rich and agglutinative language this has been a challenging task. An approach to detecting and correcting grammatical mistakes due to subject and finite-verb disagreement with regard to person, number and/or gender and due to disagreement in tense aspects in Tamil sentences is proposed in this paper. A method has been proposed that uses hierarchical part-of-speech tags of words to detect the grammatical mistakes in subject and finite-verb agreement and mistakes in tense aspects in Tamil sentences. Two sets of Tamil grammar rules are used to generate suggestions for the grammatical mistakes. Test results show that the proposed grammatical mistake detection and correction system performs well.