Automatic Myanmar News Classification

Abstract

Text classification is one of the major tasks of natural language processing and included in the interesting research areas of text data mining, which is about looking for patterns in natural language text. This paper applies two well-known classification algorithms. Algorithms applied are Naïve Bayes and k-Nearest Neighbors (KNN). These well-known algorithms are applied on collected Myanmar News dataset. Dataset used consists from 1200 documents belongs to 4 categories. The goal of text classification is to classify documents into a certain number of pre-defined categories. News corpus is used for training and testing purpose of the classifier. Feature selection algorithm is used in the proposed system to select the most relevant features from training data. Results show that precision and recall values using k-NN is better than Naïve Bayes. This research makes a comparative study between mentioned algorithms.