• Dr. SK.Syed Basha
• K. Bala vijaykumar
• CH. Vinay
• N. Bhanu mohan
• G. Rohan

Abstract—Malware and ransomware attacks have been getting more advanced leaving behind immense financial and data loss. Conventional machine learning-based detection systems primar- ily concentrate on an after execution analysis, which leads to late detection of a damage that has already been done. Moreover, the available literature is mostly based on offline data, un- explainable, and prone to evasion attacks. The following paper suggests a real-time, explainable, and resource-optimal machine learning system to detect ransomware and malware in the first stage. The suggested system constantly monitors the activity of the system, identifies threats prior to the start of file encryption, also offers interpretable information to security personnel, and is resistant to evasion. Experimental comparison shows that experimental analysis has a better ability to identify early and minimize false positives than the existing methods [4], [13] and better real-world applicability.

Malware detection, ransomware detection, machine learning, real-time monitoring, explainable artificial intelligence (XAI), early-stage threat detection, system activity monitoring, file encryption prevention, resource-optimal model, evasion attack resistance, interpretable security analysis, experimental comparison, false positive reduction, early threat identification, and real-world applicability.

"ZERO DAY MALWARE AND RANSOMWARE DETECTION", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.13, Issue 3, page no.c595-c599, March-2026, Available :http://www.jetir.org/papers/JETIR2603276.pdf

"ZERO DAY MALWARE AND RANSOMWARE DETECTION", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.13, Issue 3, page no. ppc595-c599, March-2026, Available at : http://www.jetir.org/papers/JETIR2603276.pdf