An Efficient Deep Learning Powered Real-Time Malware Identification Scheme Using Optimized Threat Classification Principle

Currently, malware is one of the most significant challenges that internet users are confronted with. This new form of malicious software is more adaptable than previous generations of infections. A generic malware is constantly modifying its signature characteristics in order to evade detection by conventional signature-based malware detection models. A variety of deep learning and machine learning techniques were employed to detect malicious threats or malware. The algorithm with the highest accuracy was chosen for implementation in the system, as evidenced by a high detection ratio. In this paper, we introduce the concepts that underlie our framework by employing an intensive learning technique known as the Optimized Threat Classification Principle (OTCP). This technique is cross-validated with the conventional learning scheme, Support Vector Machine (SVM), to assess the efficacy of the proposed scheme. This proposed learning methodology, OTCP, enhances the automated detection and classification of malware variants by constructing threat prevention networks with a greater number of potentially diverse layers, resulting in superior classification. The concepts underlying this framework were subjected to a scaling-up process after being effectively tested on medium-sized datasets of malware and clean files. This process enables us to work with huge quantities of malware and disinfect files.