A Framework for Identifying Pali Words in Myanmar Text

Abstract

Myanmar language has long been influenced by the usage of Pali words since ancient times. This study proposes a framework for identifying Pali words in Myanmar text to analyze the influence of Pali words on Myanmar language. The framework applies a combination of rulebased and dictionary look-up approach to detect Pali words written in Myanmar text. The system proposes a Pali word inventory process for collecting Pali words from real-world Myanmar online articles. After detecting the Pali words, Part-of-Speeches (POS) and corresponding meanings can be provided with the help of a Pali-Myanmar dictionary. By being able to detect Pali words in Myanmar text, the findings of this study are expected to benefit statistical analysis of Pali words influencing on Myanmar language and this, in turn, will improve further NLP tasks of Myanmar language such as spelling checking, machine translation etc.