|
Myanmar (Burmese) String Similarity Measures based on Phoneme Similarity |
|---|---|
| รหัสดีโอไอ | |
| Creator | Khaing Hsu Wai |
| Title | Myanmar (Burmese) String Similarity Measures based on Phoneme Similarity |
| Contributor | Ye Kyaw Thu, Hnin Aye Thant, Swe Zin Moe, Thepchai Supnithi |
| Publisher | Sirindhorn International Institute of Technology, Bangkadi Campus (SIIT-BKD) |
| Publication Year | 2563 |
| Journal Title | Journal of Intelligent Informatics and Smart Technology |
| Journal Vol. | 4 |
| Page no. | 27-34 |
| Keyword | Myanmar character, Burmese, String similarity metrics, Phonetic Similarity, Grapheme-to-Phoneme (G2P), Ripple Down Rules-Based (RDR) |
| URL Website | https://ph05.tci-thaijo.org/index.php/JIIST |
| Website title | Journal of Intelligent Informatics and Smart Technology |
| ISSN | 2586-9167 |
| Abstract | String similarity measurement is useful for a wide range of applications. It performs an important role in machine learning, information retrieval, natural language processing, error encoding, and bioinformatics. Measuring string similarity is also a basic and fundamental operation of data science, important for data cleaning and integration. Applications such as spell checking, duplicate finding, searching similar words, and retrieving tasks use string similarity. Moreover, Grapheme-to-Phoneme (G2P) conversion is the necessary task of predicting the pronunciation of a word given its graphemic or written form. In this study, string similarity metrics have been calculated for Burmese (Myanmar language) based on phoneme similarity and phonetic similarity. Similarity distance is measured between the datasets and query words, both of which are converted with G2P model and with the phonetic encoding mapping tables. As previous string similarity approaches are not suitable for fuzzy string matching of tonal-based Burmese, measuring string similarity based on phoneme similarity and phonetic mapping approaches are proposed in this study. |
