New Techniques in Thai-English Transliterated Words Searching - Applied to Our New Webservices Platform for Tourism (WICHAI)

Aug 28, 2026

Upload by Maeve

The research introduces a technique for Thai-English transliterated word searching using a similar-sounding words database and similarity/relatedness computations. The proposed method is integrated as the core search capability of the WICHAI tourism webservices platform to help travelers who do not speak Thai locate relevant and accurate Internet keywords with greater flexibility. Comparisons are performed with consonants and vowels mapped via RTGS, iteratively matching similar characters and using four algorithms with a similarity threshold of 0.6 to balance match quality and word-meaning understanding.

LanguageEnglish

CategoryTechnology

FormatPDF

Pages8

Reading Time20 min

AccessFree to download

Introduction

Transliteration challenges for tourism search

Transliteration for Thai Language

Consonant and vowel mapping via RTGS

Transliteration by character-level correspondence

Journal of Computers New Techniques in Thai-English Transliterated Words Searching, Applied to Our New Webservices Platform for Tourism (WICHAI) Chochiang K., Hanna F., Betbeder M. L., Lapayre J. C.* CNRS Femto-ST Institute – UMR 6174 University of France-Comte, Besancon, France. * Corresponding author. Tel.: +33.81.66.65.15; email: [jc.lapayre@femto-st.fr](jc.lapayre@femto-st.fr)[ ](jc.lapayre@femto-st.fr)[Manuscript submitted December 24](Manuscript submitted December 24), 2015; accepted May 11, 2016. doi: 10.17706/jcp.12.5.408-415 Abstract: This research proposes a new technique in Thai-English transliterated words searching by using the similar-sounding words database with similarity and relatedness calculations. The proposed search engine is used at the core of the “WICHAI” platform for tourism. The newly proposed method can help tourists, who do not speak the Thai language, to find relevant and accurate keywords on the Internet with better flexibility. The calculations use, successively, consonants and vowels comparisons based on the Royal Thai General System of Transcription (RTGS) and iterations with similar characters. Similarity calculations using 4 algorithms with a threshold of 0.6 were computed in order to obtain the most appropriate words while maintaining the level of word-meaning understanding. Key words: Thai language, transliteration, ontology, similarity. 1. Introduction Tourism is an important business section in many countries including Thailand. One of the major obstacles for tourists who wish to find information about a desired destination on the Internet is the language barrier. Some words, phrases and accents are only used in some specific social situations and in regions that have their own local dialect. Tourists may get some suggestions for unseen destinations from the locals and might not be able to find further information from the Internet. This is due to language barriers and particularly to the variation of sounds and pronunciation in the native language. For example, when tourists hear from locals about “Namtok Kathu”, which is a famous waterfall in Phuket, they will search for it on the Internet using their Latin alphabet keyboards. Tourists will most likely search for“Namtork Kathu”,“Numtoc Katu”,“Numtok Kratuu” or “Narm tog Kathu”, depending on their transliteration capabilities, instead of searching for the correct word “Namtok Kathu”. Therefore, they will not be able to find the desired destination. The new proposed algorithm helps solving this type of problems by enhancing the system’s transliterated words searching techniques. In the globalization era, most people can connect to each other via Internet anytime and anywhere. Most search engines use keywords query, which is identical and relevant to conceptual search [1]. This means that the way the user spells a word affects the search results and thus a spelling mismatch would give inaccurate results. Unfortunately, keywords for touristic attractions in Thailand are often expressed in Thai language such as the name of touristic sites, museums, restaurants or plaza. These keywords cannot be translated into English such as “วัดจันทร์” (“Wat Chan”), which is used by locals to indicate the “Chan temple”. 408 Volume 12, Number 5, September 2017 Journal of Computers When translated to English, the keyword “Wat Chan” becomes “Moon Temple” and when a tourist searches the translation he will not be able to find the desired results. Therefore, tourists have to search only in transliterated keywords. Currently, word transliteration from Thai to English has fixed guidelines and rules but, due to the language barrier, these guidelines cannot be used by tourists (because they cannot speak nor understand the Thai language) . Thereby, tourists may not find suitable and accurate information on touristic attractions on the Internet without the help of a personal travel guide or native Thai speakers. Our algorithm enhances th

FAQs

What problem does the proposed technique address for tourism searches?

It addresses the language barrier that prevents tourists from finding destination information when Thai place names and keywords are misspelled or only available through transliteration rather than direct Thai queries.

How does the method compute similarity between Thai-English transliterated words?

It compares consonants and vowels based on RTGS and performs iterative matching using similar characters. Four similarity algorithms are used with a threshold of 0.6 to select appropriate words while preserving meaning understanding.

How is the technique used in the WICHAI platform?

The algorithm serves as the core transliterated-word search engine in the WICHAI tourism webservices platform, enabling more flexible and accurate keyword retrieval for travelers who cannot read Thai.

Aug 28, 2026

Upload by Maeve