Malware Detection Using Machine Learning

Malware Detection Using Machine Learning

Somesh Paraganve, Prathmesh Kumbhar , Prasanna Chougule , Pratik Patil , Shubham Zambre

Prof .Shruti Narde

Dr. J. J. Magdum college of Engineering, Jaysingpur.

Abstract- Malware,a persistent threat to cybersecurity,    continuously evolves, rendering traditional detection methods ineffective. This paper presents a novel machine learning approach to malware detection, addressing the shortcomings of signature-based systems. Leveraging AI and statistical models, our system autonomously identifies malware, including polymorphic and zero-day threats, by learning from extensive datasets. This proactive defense mechanism enhances computer system security amid the digital landscape's complexities.our method combines multiple machine learning models, recognizing both known signatures and emerging patterns of malicious behavior. By analyzing vast datasets, our system achieves high accuracy with minimal false positives, surpassing traditional techniques. As millions of new malware samples emerge daily, adaptive detection methods are imperative. Our research combines dynamic and static features, achieving promising results across various malware types, contributing significantly to cybersecurity advancements. Malware encompasses diverse programs aimed at harming computer systems, networks, or servers.