Abstract:
In Myanmar language, the spoken
language may be different with the written
language in many cases. In Myanmar, the words
are the same phonetically in pronunciations with
different spellings such as ကုနရန်, ယန်ဂုန်,
ရွှေတိဂုန်,ရွှေဒဂုန်. Phonetic string matching is a
research area of information retrieved systems,
search systems, data mining, text and web-mining,
speech recognition and pattern recognition
systems.The purpose of this paper is to present
phonetic matching algorithm in Myanmar
language. In this paper, we use Myanmar
character-based NE transliteration algorithm
based on Romanized words. This transliteration
includes three steps: Normalization, Word
Splitting, and Transliteration. Firstly, we
normalized Myanmar input words. And then, we
transliterated these Myanmar names with
Romanized Myanmar consonants and vowels to
English. Finally, Metaphone algorithm is applied
to transliterated words in finding similarities of
words based on language pronunciation.
