PERBANDINGAN METODE NAÏVE BAYES DAN SVM UNTUK SENTIMEN ANALISIS MASYARAKAT TERHADAP SERANGAN RANSOMWARE PADA DATA KIP-K

Abstract

This research examines ransomware attacks on KIP-K data by analyzing the opinions of Social Media X users, using the naïve bayes classifier (NBC) and support vector machine (SVM) methods. The rapid development of technology not only brings great benefits but also increases the risk of digital attacks by certain parties. One example is a ransomware attack that caused a KIP-K data leak. In this study, sentiment analysis was applied to identify public opinions or responses obtained from Social Media X, with the help of python programming and google colab. Of the total 2,648 raw data collected, pre-processing was carried out resulting in 1,738 cleaned data. The study compared two methods, namely Naïve Bayes and Support Vector Machine, to determine what method is more effective in analyzing public sentiment related to ransomware attacks on KIP-K data. The focus of this study is to understand the percentage of Social Media X users' comments and responses related to the KIP-K ransomware taken from media sosial X. The stages of sentiment analysis in this study include crawling, labeling, preprocessing, method classification, and visualization. Before the classification process was carried out, the data was divided into two parts, namely 30% for test data and 70% for training data. Data labeling resulted in 1,313 negative data, 957 positive data and 377 neutral data. The classification results show that the NBC method has an accuracy of 70%, while the SVsM achieves an accuracy of 88%. Based on these results, SVM is proven to be superior in data analysis compared to NBC, especially for big data.