An Novel Approach in Malware Detection with Cryptography Based Approach Using Machine Learning

Résumé fourni par la source

A critical concern that arises these days is identifying and controlling the advancement of malware threats. Recognizing malware through conventional anti-malware tools has become challenging due to its constant development. The constant emergence of malware renders common malware-identifying methods less effective, necessitating the use of advanced artificial intelligence in the form of machine learning models to identify and mitigate the spread of malware. The field of artificial intelligence (AI) has resulted in major developments in technology. These advancements have had an essential influence on cryptography, where AI methodologies exhibit outstanding potential for improvising the accuracy and stability of the cryptographic system. Malware frequently uses cryptographic methods to conceal its malicious activities and hidden features from antivirus engines, making it easier for security professionals to identify and revert it. Recognition of cryptographic functions in malware becomes crucial in detecting and precisely examining malicious content. Numerous efforts have been taken for tackling the existing problem, but the methodologies currently in use have major disadvantages, including cost prohibitive restrictions due to prior knowledge and failure in the achievement of results with high accuracy. This paper provides a comprehensive overview of malware detection using AI and a cryptographic approach. We utilized machine learning models like random forest and blowfish cryptography to identify malware data, achieving an accuracy of 88.24%. This model outperformed traditional machine learning models like KNN 54.13, ANN 56.82, and Gradient Boosting 68.21 in terms of accuracy.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.