Classifi cation of Thai Independent Study in Statistics Using Data Mining Techniques

บทคัดย่อ

Abstract

In this paper, the empirical study of the classification of Thai independent study in statistics is discussed. Our purpose is to classify the undergraduate independent study researches into three groups: sample survey, statistical analysis,and operational research and related field. Several classification techniques, such as support vector machine, Naïve Bayesian, Decision Tree, k-Nearest Neighbor and RBF network, are used in this paper. We also employed the feature selection techniques in order to find the best subset of features that help improve the accuracy of the classification model. The experimental results show that the RBF network algorithm gives a best accuracy when the Chi-square is employed as the feature selection method.

Keywords: Document Classification, Independent Study, Data Mining, Text Mining