KLASIFIKASI KATEGORI WEBSITE MENGGUNAKAN ALGORITMA NAÏVE BAYES DAN SVM

Abstract

With the explosion of information in the 21st century, information overload has become a major challenge on the internet. Although search engines help users assess the value of websites based on their topics, finding specific information remains challenging. This study proposes a practical approach by creating automatic data records that summarize the content of each website for categorization purposes. Text classification (TC) tasks automatically assign documents to specific categories. This study also focuses on text classification methods, including Naïve Bayes and Support Vector Machine (SVM), to achieve good accuracy. In addition to these two methods, this study details the use of other methods such as K-Nearest Neighbors (KNN) and Random Forest in the context of web phishing classification and sentiment analysis. In this experiment, Naïve Bayes and SVM were evaluated for classifying website categories. The data was divided into training (70%) and testing (30%) with accuracy results of around 88% for Naïve Bayes and 85% for SVM. This study provides a deeper understanding of the performance of two different classification methods in the context of website categorization.