Examinations on the Performance of Classification Models for Thai News Articles

2019 11th International Conference on Information Technology and Electrical Engineering (ICITEE) Pub Date : 2019-10-01 DOI:10.1109/ICITEED.2019.8929959

Arisara Noppakaow, O. Uchida

引用次数: 2

Abstract

This research aims to examine automatic models to classify Thai online news articles. The data set is six thousands of news articles from three mainstream websites. The news articles are classified into four categories—crime news, politic news, sport news, and entertainment news. Examinations on the classification algorithms of Decision Tree, Support Vector Machine (SVM), and Deep Learning are conducted. The performance is measured by the accuracy, the recall, the precision, and the F-Measure. The results show that the accuracies of Decision Tree, SVM, and Deep Learning models are 86%, 94%, and 95%, respectively.

查看原文本刊更多论文

泰国新闻文章分类模型的性能检验

本研究旨在检验自动模型对泰国在线新闻文章进行分类。该数据集是来自三个主流网站的6000篇新闻文章。新闻文章被分为四类——犯罪新闻、政治新闻、体育新闻和娱乐新闻。对决策树、支持向量机(SVM)和深度学习的分类算法进行了检验。性能由准确性、召回率、精度和F-Measure来衡量。结果表明，决策树、支持向量机和深度学习模型的准确率分别为86%、94%和95%。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

2019 11th International Conference on Information Technology and Electrical Engineering (ICITEE)

自引率

0.00%

发文量