Detection of malware using machine learning methods

Malware is software designed to damage computer systems, steal data or render systems unusable. Today, detecting and preventing malware is of great importance in the field of cyber security. In this thesis study, malware detection and analysis was carried out with machine learning-based algorithms. In the thesis study, the KDD-cup-99 dataset, which is a 10% subset of the original KDD99 dataset, was used. The dataset contains 42 features and 494,020 records. The target feature specifies the connection type, which can be classified into normal and abnormal connection types. But abnormal connections can also be divided into four types of cyberattacks namely Probe, Denial of Service (DOS), R2L (Root 2 Local) and U2R (User 2 Root). Five different machine learning algorithms, including logistic regression algorithm, ensemble algorithm, artificial neural networks, support vector machines and k-nearest neighbor algorithms, were run in attack detection and their results were compared. Artificial Neural Networks (ANN) model showed the best performance with an accuracy rate of 97.61%. The results obtained show that machine learning methods have high accuracy rates in malware detection. These findings will contribute to the development of more effective detection and prevention strategies in the field of cybersecurity. KEYWORDS: Machine Learning, Malware, Classification