Pencocokan Pelanggaran UU ITE dalam Kejahatan Siber Menggunakan Algoritma Support Vector Machine

Abstract

The process of identifying and categorizing cyber law violations in Indonesia has traditionally involved manual procedures managed by experts, lawyers, or law enforcement. Reliance on expert witnesses to identify violations of the ITE Law in cybercrime presents significant challenges due to limited discretion and high costs. The emergence of information technology undeniably brings various benefits to society. However, this development is not without its drawbacks as it has inadvertently given rise to unethical behavior, known as cybercrime. Users can use this web-based application to find out the article of the ITE Law that was violated from the chronology of cybercrime cases. This research aims to match the chronology of cybercrime with the article of violation of the ITE Law using the Support Vector Machine method and the Radial Basis Function Kernel to automate classification and improve the analysis and processing of chronological data of cybercrime cases through the application of text mining. The data used in this study amounted to 617 data consisting of 493 training data and 124 test data by scrapping data on one of the websites covering the chronology of cybercrime in Indonesia. The model built obtained an accuracy rate of 91% based on the classification of articles of violation of the ITE Law in Indonesia by applying the confusion matrix evaluation method.