The Analysis and Summarizing System of Thai Hotel Reviews Using Opinion Mining Technique

被引:4
作者
Sungsri, Teerapong [1 ]
Ua-apisitwong, Usanad [1 ]
机构
[1] 340 Nakhonratchasima Univ, Fac Sci & Technol, Nakhon Ratchasima, Nakhonratchasim, Thailand
来源
ICIET'17: PROCEEDINGS OF THE 5TH INTERNATIONAL CONFERENCE ON INFORMATION AND EDUCATION TECHNOLOGY | 2017年
关键词
Opinion Mining; Opinion feature; Polarity classification; Hotel review summarization;
D O I
10.1145/3029387.3029391
中图分类号
TP39 [计算机的应用];
学科分类号
081203 ; 0835 ;
摘要
This paper proposes a development of a new opinion mining framework for analyzing and summarizing the hotel reviews on Thai language. This framework uses a hybrid concept between a keyword knowledge and a classification technique to identify a opinion feature of hotel reviews into 3 aspects (location, service and worthiness) and to classify the opinion polarity of hotel reviews into 2 classes (positive and negative). The research methodology consists of 4 steps as 1) preprocessing 2) hotel feature identification 3) opinion polarity classification and 4) hotel review summarization. Experiments showed that the accuracy of the opinion feature identification is 83.33% and the accuracy of the opinion polarity classification is 81.47% with a testing data which be manually collected from a hotel review on the agoda website.
引用
收藏
页码:167 / 170
页数:4
相关论文
共 10 条
  • [1] Alkadril A.M, 2016, INT J ADV COMPUTER S, V7
  • [2] [Anonymous], 2011, P 4 ACM INT C WEB SE, DOI DOI 10.1145/1935826.1935884
  • [3] [Anonymous], P C KNOWL DISC DAT M
  • [4] [Anonymous], 2006, P ACM INT C INF KNOW
  • [5] Farra N., 2010, Proceedings 2010 10th IEEE International Conference on Data Mining Workshops (ICDMW 2010), P1114, DOI 10.1109/ICDMW.2010.95
  • [6] HARUECHAIYASAK C., 2010, Proceedings of the 8th Workshop on Asian Language Resouces, Beijing, P64
  • [7] Jin W, 2009, P C KNOWL DISC DAT M
  • [8] Liu B, 2010, CH CRC MACH LEARN PA, P627
  • [9] Sukhum K, 2011, INFORM TECHNOLOGY J, V7, P32
  • [10] Yessenalina Ainur, 2010, P 2010 C EMP METH NA, P1046