Survey on Thai NLP Language Resources and Tools

Proceedings of the 13th Conference on Language Resources and Evaluation (LREC 2022), pages 6495–6505 Marseille, 20-25 June 2022 © European Language Resources Association (ELRA), licensed under CC-BY-NC-4.0 6495 Survey on Thai NLP Language Resources and Tools Ratchakrit Arreerard, Stephen Mander, Scott Piao School of Computing and Communications Lancaster University Lancaster LA1 4WA {r.arreerard,s.mander3,s.piao}@lancaster.ac.uk Abstract Over the past decades, Natural Language Processing (NLP) research has been expanding to cover more languages. Recently particularly, NLP community has paid increasing attention to under-resourced languages. However, there are still many languages for which NLP research is limited in terms of both language resources and software tools. Thai language is one of the under-resourced languages in the NLP domain, although it is spoken by nearly 70 million people globally. In this paper, we report on our survey on the past development of Thai NLP research to help understand its current state and future research directions. Our survey shows that, although Thai NLP community has achieved a significant achievement over the past three decades, particularly on NLP upstream tasks such as tokenisation, research on downstream tasks such as syntactic parsing and semantic analysis is still limited. But we foresee that Thai NLP research will advance rapidly as richer Thai language resources and more robust NLP techniques become available. Keywords: Natural Language Processing, Thai NLP, Survey, Language Resource, NLP tools 1. Introduction Over the past decades, Natural Language Processing (NLP) has advanced rapidly, having become a major research area with a wide range of applications in AI (Artificial Intelligence) and ICT (Information Commu- nication Technology) systems. Today, NLP language resource and tools such as WordNet (Fellbaum, 1998), CoreNLP (Manning et al., 2014) and NLTK (Loper and Bird, 2002) cover numerous languages, and keep ex- panding to include more languages. However, there are still many languages for which more language re- sources and NLP tools need to be developed. In partic- ular, unique features of some languages demand specif- ically tailored NLP techniques and approaches. Thai language is one of the under-resourced languages in the NLP domain, although it is spoken by approximately 70 million people globally1. In fact, Thai NLP research has a long history, started about three decades ago (Sornlertlamvanich, 2019). The earliest published Thai NLP work we could trace back includes an automatic Thai syllabus analysis of Poowarawan (1986) and Multi-Lingual Machine Trans- lation (MMT) Project (Funaki, 1993). Since then, nu- merous NLP language resources and tools have been developed and published for Thai language, such as TLTK (Aroonmanakun, 2002) and AIforThai plat- form2. Despite recent significant development of NLP re- search in Thai NLP community, however, compared to some other major languages such as English, there is still much room for further development. In partic- ular, Thai language has some unique features, for in- stance Thai writing system lacks explicit delimiter of words (similar with Chinese and Japanese languages), and Thai syntactic grammar allows all Subject-Verb- 1https://en.wikipedia.org/wiki/Thailanguage 2See https://aiforthai.in.th/ Figure 1: example Thai sentences Object, Subject-Object-Verb and Object-Subject-Verb structures, as shown by Figure 1 …

Content truncated to 500 words