Comparative Analysis of LSTM and CNN for Efficient Malware Detection
Frances Osamor, Briana Lowe Wellman · 2022
Intrusion-based Detection Systems are recognized as a crucial element in the safety of an organization's network infrastructure. It is the base responsible for detecting any potential threat. To detect different attacks, many IDS tools incorporate the analysis of the network traffic using the flow-based network. Research in network safety is a developing field, and IDS-specific is the current emphasis, with numerous studies and methods developed and proposed. In this study, we propose and compare the use case of a deep learning model, Long Short-Term Memory, and Convolutional Neural Network for efficient malware detection. We use the open-source benchmark dataset, ADFA-LD, for training and assessment. The experimental analysis shows that the proposed models can achieve promising results in detecting the malware concerning the recall, accuracy, and false positive rate.