Punyaphol Horata. A study of sentiment analysis using long short term memory techniques on Thai twitter data. (). King Mongkut's University of Technology North Bangkok. Central Library. : , 2023.
A study of sentiment analysis using long short term memory techniques on Thai twitter data
Abstract:
Sentiment analysis studies are now very important and useful in business to be able to recognize social customers. Especially when there is a Twitter user, which in Thailand has a relatively large number of users who use Twitter. Almost all previous works used basically classify techniques such as SVM, Naïve Bayes, etc., deep learning techniques. Accuracy in this domain has been demonstrated in a corpus of English tweets. In this paper, we present a study using deep learning techniques to classify the sentiment of Thai Twitter data. Long Short-Term Memory (LSTM) was used in this study. We also examined the impact of word order on Thai tweets. The model was able to categorize the sentiments in the test to be either positive or negative with an approximate accuracy of 83.04%. These accuracy values can be further improved by using different neural network architectures and/or data pre-processing and parameter augmentation.
King Mongkut's University of Technology North Bangkok. Central Library