Experiments on the Use of Feature Selection and Machine Learning Methods in Automatic Malay Text Categorization

Due to the rapid growth of documents in digital form, research in automatic text categorization into predefined categories has witnessed a booming interest. Although, there is a wide range of supervised machine learning methods have been applied to categorize English, relatively, only a few studies have been done on Malay text categorization. This paper reports our comparative evaluation of three machine learning methods on Malay text categorization. Two feature selection methods (Information gain (IG) and Chi-square) and three machine learning methods (K-Nearest Neighbor (k-NN), Naive Bayes (NB) and N-gram) were investigated. The three supervised machine learning models were evaluated on categorized Malay corpus, and experimental results showed that the k-NN with the Chi-square feature selection gave the best performance (Macro-F1 = 96.14). (C) 2013 The Authors. Published by Elsevier Ltd.