Spelling Checking and Error Corrector System for Marathi Language Text Using Minimum Edit Distance Algorithm

The performance of any search engine, social media word processor depends deeply on the spelling checkers, grammar checkers etc. Spelling checker is the application used to correct the spelling mistakes done by users unintentionally. The minimum edit distance is one of the string-matching algorithms used in various applications like text mining, spell checking, bioinformatics and so on. In this paper, we proposed the minimum edit distance algorithm (MED) which correct the spelling mistakes in Marathi language text. It corrects the spelling errors by performing various operations like substitution, insertion, and deletion of characters. The algorithm detects non-word spelling errors and generates a suitable suggestion set for misspelled words by matching them with the corpus. While doing this, the misspelled word length is a key component to searching in the corpus so that, the searching complexity is minimum. In this paper, we have implemented a minimum edit distance algorithm and evaluated their performance with accuracy measure. The accuracy of the system is 85.5%. The performance of this algorithm is evaluated by suggestion generation accuracy for given misspelled words.