Title :
Spell checker for Myanmar language
Author_Institution :
Univ. of Comput. Studies, Mandalay, Malaysia
Abstract :
Natural Language Processing (NLP) is one of the most important research areas carried out in the world of Human language. For every language, spell checker is an essential component of many of the common Desktop applications, Machine Translation system and Office Automation system. In Myanmar, Myanmar Language is used as an official language. Myanmar Pronunciation and orthography has differences because spelling is often not an accurate reflection of pronunciation. In this paper, we developed Myanmar Spell Checker which can handle Typographic Errors (Non-word Errors), Phonetic Errors and Sequence Errors of Myanmar words. If misspelled word contains in the input sentence, this system can provide suggestion for misspelled Myanmar words. We apply Myanmar text Corpus to check Myanmar words. And then we used String Cosine Similarity to generate suggestions list for mistyped Myanmar words. The system can improve the quality of suggestion for misspelled Myanmar words and users´ efficiency when the users cannot figure out the correct spelling by themselves.
Keywords :
linguistics; natural language processing; text analysis; Myanmar language; Myanmar orthography; Myanmar pronunciation; Myanmar spell checker; Myanmar text corpus; NLP; desktop applications; human language; machine translation system; misspelled Myanmar words; natural language processing; nonword errors; office automation system; official language; phonetic errors; sequence errors; string cosine similarity error; suggestion list generation; typographic errors; user efficiency; Automata; Conferences; Context; Dictionaries; Educational institutions; Encoding; Natural language processing; Myanmar spell checker; natural language processing; string cosine similarity; text corpus; tokenization;
Conference_Titel :
Information Retrieval & Knowledge Management (CAMP), 2012 International Conference on
Conference_Location :
Kuala Lumpur
Print_ISBN :
978-1-4673-1091-8
DOI :
10.1109/InfRKM.2012.6204974