Machine Learning Techniques for Malware Detection with Challenges and Future Directions

Mohammed A Alqhatani(1*)
(1) Imam Abdulrahman Bin Faisal University
(*) Corresponding Author
DOI : 10.54039/ijcnis.v13i2.5047


In the recent times Cybersecurity is the hot research topic because of its sensitivity. Especially at the times of digital world where everything is now transformed into digital medium. All the critical transactions are being carried out online with internet applications. Malware is an important issue which has the capability of stealing the privacy and funds from an ordinary person who is doing sensitive transactions through his mobile device. Researchers in the current time are striving to develop efficient techniques to detect these kinds of attacks. Not only individuals are getting offended even the governments are getting effected by these kinds of attacks and losing big amount of funds. In this work various Artificial intelligent and machine learning techniques are discussed which were implements for the detection of malware. Traditional machine learning techniques like Decision tree, K-Nearest Neighbor and Support vector machine and further to advanced machine learning techniques like Artificial neural network and convolution neural network are discussed. Among the discussed techniques, the work got the highest accuracy is 99% followed by 98.422%, 97.3% and 96% where the authors have implemented package-level API calls as feature, followed by advanced classification technique. Also, dataset details are discussed and listed which were used for the experimentation of malware detection, among the many dataset DREBIN had the most significant number of samples with 123453 Benign samples and 5560 Malware samples. Finally, open challenges are listed, and the future directions are highlighted which would encourage a new researcher to adopt this field of research and solve these open challenges with the help of future direction details provided in this paper. The paper is concluded with the limitation and conclusion section

Article Statistic

Abstract view : 119 times
PDF views : 50 times

How To Cite This :


  • There are currently no refbacks.

International Journal of Communication Networks and Information Security (IJCNIS)               ISSN: 2073-607X (Online)