TY  - JOUR
T1  - Malware Analysis and Detection Approaches: Drive to Deep Learning
AU - Ali, Toqeer AU - Jan, Salman AU - Niza Musa, Shahrul AU - Rahman, Atiqur 
JO  - Journal of Engineering and Applied Sciences
VL  - 13
IS  - 15
SP  - 6281
EP  - 6292
PY  - 2018
DA  - 2001/08/19
SN  - 1816-949x
DO  - jeasci.2018.6281.6292
UR  - https://makhillpublications.co/view-article.php?doi=jeasci.2018.6281.6292
KW  - Trusted computing
KW  -neural networks
KW  -RNN
KW  -ESN
KW  -CNN
KW  -GAN
KW  -DCGAN
KW  -GPU
AB  - The growing number of malware attacks poses serious threats to private data and to the expensive
computing resources. To detect malware and their associated families, anti-malware companies rely on
signatures which indeed include regular expressions and strings. The recent malware attacks in the last few
years including the resurgence of ransomware have proven that signature-based methods are error-prone and
can be easily evaded by intelligent malware programs. This study reviews traditional and state-of-the-art models
developed for malware analysis and detection. According to our observation the classification of malware and
their behavior facilitates in provision of basic insights for the researchers working in the domain of malware
analysis. At the end we present the conception of using Deep Convolutional Generated Adversarial Networks
(DCGAN) in the area of malware detection as the DCGANs are the latest approach in deep learning that
effectively deals adversarial examples.
ER  -