Classification of Indonesian Online News Topics Using Text Mining

Abstract

Online news is in great demand by many people along with the rapid development of information technology. It needs to develop a system that can automatically classify news according to news categories using text mining method. In this research, the methods used in the classification are K-Nearest Neighbor. News classification is done to classify news into 4 categories, namely automotive, technology, money and food. Each topic contains 300 news data. Before the classification began, data in the form of texts will first be carried out in the preprocessing stage and weighted using TF-IDF. Based on the evaluation using the confusion matrix with the division of the dataset into 960 training data and 240 test data, with the value of k=7 it obtained accuracy of 93.75%, Precision of 94.09%, Recall of 93.56%, and F1-Score of 93.71%.