Klasifikasi Berita Berdasarkan Kategori Menggunakan Multinomial Naïve Bayes dengan K-Cross Validation dan Seleksi Fitur Chi-Squared

Bahasa Indonesia

Abstract

Classifying news articles based on categories is an important challenge in text analysis and natural language processing. Most categorization of online news articles is often done manually, making it a complex and time-consuming process. To address this issue, the development of an automatic system capable of classifying news articles into various categories such as technology, sports, and entertainment is needed. The system is built using an approach to classify news articles into several appropriate categories using the Naïve Bayes method with TF-IDF weighting and feature selection using Chi-Squared. The Naïve Bayes model training uses the reduced feature results of 10,000 features from 54,091 features. Evaluation results show that the Naïve Bayes approach is able to produce a news classification model with good accuracy, with accuracy, precision, recall, and f1-score values of 96%.