A Robust Bangla Spell Checker for Search Engine
Md. Mijanur Rahman, Hasan Mahmud, Razia Sultana Rupa, Mahnuma Rahman Rinty · 2021
The spell-checking system assists to detect spelling-errors and advise the appropriate corrections in the word documents automatically. This paper is concerned with the implementation of an interactive Bangla spell checking tool, and demonstrating how to use it in a general-purpose search engine effectively. From the literature, it is seen that Bangla has complex grammatical and orthographical rules for spelling, and the study addressed different challenges in generating suggestions for Bangla phonetic-errors. The study also presented a combined technique of the string-matching algorithm and a reverse dictionary lookup method for detecting the typographical-errors as well as cognitive phonetic-errors in Bangla language. Both algorithms have been implemented into a search engine app for locating a misspelled word, and listing the suggested candidate-words. Finally, the intended word is chosen from the candidate words that were generated by the spell checker, and used it for searching in the search engine platform. Two-level of databases (the lexicon and secondary database) were used to design the proposed system. The lexicon database contains about 58,319 correctly spelled Bangla words, and a secondary database contains about 46,881 stemmed-key data.