A novel hybrid approach to detect and correct spelling in Tamil text

Abstract

A spell checker is a tool to check the spelling of words in a document, validate them and in case of finding error in spelling, list out the possible words as suggestions. There are several researches available on spell checkers for European languages and Indian languages. But few for Tamil language, maybe, because of its nature as it is a highly inflectional and complex language. An efficient error detection and correction approach for spell checking in Tamil language has been proposed in this paper. A novel hybrid approach is adopted using tree-based algorithm with stemming and n-gram techniques. Testing results show that our system detects perfectively the error in spelling and provides most suitable suggestions for correcting the misspelt words with minimum of 91% accuracy.