Forensic Digital Analysis of Telegram Applications Using the National Institute Of Justice and Naïve Bayes Methods

Abstract

Currently, Telegram is an instant messaging application that is often used by the Indonesian people as a means of long-distance communication with other users. Telegram also has good security features to protect all data from its users. However, Telegram has a positive impact on its users. This security feature can be used by several people to protect against digital crimes, especially cases of sexual harassment. To overcome the existing crimes, analysis, and forensic methods are needed to help solve crimes. This research is guided by the investigation process using the National Institute Of Justice (NIJ) method and the Naïve Bayes method to classify the conversations found. It can be concluded that MOBILedit Forensic Express has a poor performance in finding digital evidence in the Telegram application and FTK Imager is very good at finding digital evidence in the Telegram application. In this research, the classification process using the Naïve Bayes method has been able to classify conversations that contain sexual harassment or not. Evaluation of the classification method uses a confusion matrix to determine the best classification model.