Paper Key : IRJ************046
Author: P.jyothsna,V.l.rohith,S.samhitha,S.uma Mahesh Patnaik,B.nageswara Rao
Date Published: 04 Apr 2024
Abstract
In the last few years, there has been an increase in cybercrimes which have turned into a multi-billion-dollar industry. Most cybercrimes entail having to use a malware. Malware developers keep revising their approaches and tactics as they create attacks that are hard to detect and can remain inactive for long periods of time while bypassing security systems. Deep learning models consequently become very popular for identification and classification of malware. A uniquely deep learning ResNet model is put forward in this study for detection with an amazing accuracy on both new and old samples. It outperforms previous literature by using a deep convolutional neural network (RESNET -50). In addition, important aspects like malware detection, image features, and malimg dataset, malware visualization have attention to the practical method of addressing convoluted variations. This paper demonstrates how deep learning can be effectively used to identify malware.Keywords: deep learning, malware detection, ResNet-50, image features, malware visualization, malimg dataset.