A Hybrid Approach For Malware Classification Using Secondary Features Fusion
Researchers have developed a novel hybrid approach for malware classification, aiming to improve detection and mitigation efforts. The method fuses secondary features, including API calls and n-grams, using a customized selection process. A voting-based algorithm fusion is employed for the predictive model, achieving high accuracy with an AUC of 0.989 and 99.72% on a Microsoft dataset. AI
IMPACT This novel approach could enhance cybersecurity defenses by improving the accuracy and efficiency of malware family classification.