MALWARE DETECTION USING HONEYPOT AND MACHINE LEARNING

• Main Author: Muhamad Malik Matin, Iik
• Format: Theses
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/53886
• Institution: Institut Teknologi Bandung

Malware is one of the main threats to the security of computer users. Malware has impacts that can disrupt and harm users, computers or networks. Some of the techniques in detecting malware consist of static analysis, dynamic analysis and machine learning. Machine learning has the ability to be quite effective and efficient compared to other techniques. In its application, machine learning has limitations when it uses very little training data. Some malware datasets are not updated and therefore have the potential to increase false positives when machine learning tries to detect new malware. Therefore, a solution is needed that can update the existing dataset. Updates to the dataset can be done by added the latest malware information. A honeypot is a system that has the ability to detect and collect new malware. In this study, a detection technique for malware using a honeypot and machine learning was designed. The author combines the ClaMP dataset with the feature extraction results on the honeypot. Then the authors built a model using machine learning. At the end of this study the authors succeeded in increasing the accuracy by 0.14% with the achievement of 99.37% and a recall of 0.06% with the achievement of 99.73%.