Malware Attacks Detection in Network Security using Deep Learning Approaches
Humaira Naeem, Asma Batool
This abstract provides an overview of the study on the use of deep learning approaches, specifically Recurrent Neural Networks (RNNs) and Long Short-Term Memory Networks (LSTMs), for detecting malware attacks in network security. The increasing sophistication of malware attacks has made it challenging for traditional signature-based approaches to detect them effectively. Deep learning algorithms offer the potential to address these challenges, as they can automatically learn complex representations of the data and adapt to new and evolving threats. The study focused on the collection and analysis of a large and diverse dataset of both benign and malicious software samples, which were used to train and validate the deep-learning models. The results of the study showed that the RNN and LSTM algorithms outperformed traditional signature-based approaches in terms of accuracy and efficiency in detecting malware attacks. Additionally, developing more efficient and scalable training methods for deep learning algorithms is an important area for future research. Overall, the future of malware detection using deep learning is promising, and continued research in this field holds great potential for improving the security of our digital systems.