|
Enhancing the Performance of Sentiment Analysis Models Using GridSearchCV: A Case Study on Electric Vehicles in Thailand |
|---|---|
| รหัสดีโอไอ | |
| Creator | Chirawat Emphan |
| Title | Enhancing the Performance of Sentiment Analysis Models Using GridSearchCV: A Case Study on Electric Vehicles in Thailand |
| Contributor | Ekkasit Tiamkaew, Sanya Khruahong |
| Publisher | Faculty of Informatics, Mahasarakham University |
| Publication Year | 2569 |
| Journal Title | Journal of Applied Informatics and Technology |
| Journal Vol. | 8 |
| Journal No. | 1 |
| Page no. | 260631 |
| Keyword | Electric Vehicles, GridSearchCV, Hyperparameter Tuning, Machine Learning, Natural Language Processing, Sentiment Analysis, Tokenizing |
| URL Website | https://ph01.tci-thaijo.org/index.php/jait |
| Website title | Journal of Applied Informatics and Technology |
| ISSN | 3088-1803 |
| Abstract | This study investigates the enhancement of sentiment analysis model per-formance through hyperparameter tuning using GridSearchCV, with afocus on electric vehicle reviews in Thailand. To address the challengesinherent in Thai language processing, PyThaiNLP was employed for wordsegmentation and text preprocessing. Four machine learning models—Support Vector Machines (SVM), Multinomial Na ̈ıve Bayes, Logistic Re-gression, and Stochastic Gradient Descent—were implemented for senti-ment classification. The hyperparameters of each model were systemati-cally optimized to determine the configuration that maximizes accuracy,precision, recall, and F1-score. Experimental results revealed notable im-provements across all models following optimization. The SVM model,identified as the best-performing classifier, achieved an accuracy increasefrom 75.00% to 76.69% and an F1-score improvement from 74.82% to76.69%. The optimal SVM configuration employed a radial basis func-tion kernel with a regularization parameter (C) of 10 and a gamma valueof 0.1. These findings underscore the significance of hyperparameter op-timization in improving model effectiveness and contribute to advancingsentiment analysis in linguistically complex environments such as Thai. |
