Semi-supervised Thai Sentence Segmentation Using Local and Distant Word Representations

Saetia, Chanatip; Chuangsuwanich, Ekapol; Chalothorn, Tawunrat; Vateekul, Peerapon

doi:10.4186/ej.2021.25.6.15

Cited by 3 publications

(2 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Japanese and Chinese have no word boundaries [10], [11]. Therefore, several NLP tasks must segment sentences into words [12], [13]. Many techniques develop the existing Thai word segmentation programs.…”

Section: A Word Segmentationmentioning

confidence: 99%

Customer Needs Classification from Online Social Media Using Bag-of-Concepts Representation

Laosen,

Intana,

Chuaynukul

2023

Int. J. Adv. Sci. Eng. Inf. Technol.

View full text Add to dashboard Cite

Social media platforms are now a very powerful tool for digital marketing strategy because it helps companies to be in direct contact with their customers. A communication problem in digital social media is several customer needs phrases posted on social media, making it difficult for businesses to find relevant posts and respond to customers immediately. Therefore, knowing and understanding the customer requirements for a product can help the product owner to propose the right product to the right customer. This study focuses on understanding customer needs in Thai and classifying them into certain concepts. This study aims to classify customer needs for products in online social media community groups. The model focuses on understanding Thai customer need phrases. We then use a bag of concepts representation, including pattern analysis that applies n-grams together with POS and synonym replacement, conceptual analysis, pattern matching, and class labeling that applies concept sets obtained from the FP-Growth algorithm and represents TD-IDF value in a bag of concepts. The effectiveness of the proposed model is evaluated on five classification algorithms, including Decision Tree, Support Vector Machine, NaÃ¯ve-Bayes, K-Nearest Neighbor, and RBF Neural Network. The results show that Decision Tree can yield higher accuracy and F-measure values than the others. As this study is an initial step of a personalized product recommendation system in the future, this study will apply the model to the remaining domains for future work.

show abstract

Section: A Word Segmentationmentioning

confidence: 99%

Customer Needs Classification from Online Social Media Using Bag-of-Concepts Representation

Laosen,

Intana,

Chuaynukul

2023

Int. J. Adv. Sci. Eng. Inf. Technol.

View full text Add to dashboard Cite

show abstract

“…Some literature suggests that certain languages, such as Arabic, Lao, and Khmer, lack any clear marker for sentence boundaries (Charoensuk et al, 2005;Saetia et al, 2021). However, this claim is not entirely accurate, as these languages do employ punctuation marks to denote the end of a sentence.…”

Section: Sentence Boundaries and Punctuationmentioning

confidence: 99%

Thai sentence segmentation using large language models

Panitsrisit

View full text Add to dashboard Cite

Thai sentence segmentation has been on the topic of interest among Thai NLP communities. However, not much literature has explored the use of transformer-based large language models to tackle the issue. We conduct three experiments on the LST20 corpus, including (1) fine-tuning WangchanBERTa, a large language model pre-trained on Thai, across different classification tasks, (2) joint learning for clause and sentence segmentation, and (3) cross-lingual transfer using the multilingual model XLM-RoBERTa. Our findings show that WangchanBERTa outperforms other models in Thai sentence segmentation, and fine-tuning it with token and contextual information further improves its performance. However, cross-lingual transfer from English and Chinese to Thai is not effective for this task.

show abstract

Thai-English Neural Machine Translation Method with Local and Global Syllable Feature

Xiong,

Liu

2023

2023 International Conference on Asian Language Processing (IALP)

View full text Add to dashboard Cite

Semi-supervised Thai Sentence Segmentation Using Local and Distant Word Representations

Cited by 3 publications

References 0 publications

Customer Needs Classification from Online Social Media Using Bag-of-Concepts Representation

Customer Needs Classification from Online Social Media Using Bag-of-Concepts Representation

Thai sentence segmentation using large language models

Thai-English Neural Machine Translation Method with Local and Global Syllable Feature

Contact Info

Product

Resources

About