Detection of Malware Using Machine Learning techniques on Random Forest, Decision Tree, KNeighbour, AdaBoost, SGD, ExtraTrees and GaussianNB Classifier
Kanwarpartap Singh Gill, Vatsala Anand, Rupesh Gupta, Pao‐Ann Hsiung · 2023
Spotting harmful software with machine learning models has become more crucial in cybersecurity. Machine learning can be used to recognize and sort harmful software by looking for specific patterns, behaviours, and features that set them apart from normal programs. It’s important to understand that the world’s access of harmful computer programs is always changing, and people who create these programs often try different methods to avoid getting caught. So, updating the model regularly and keeping an eye on it all the time are very important to making sure we can detect malware effectively. In the current digital world, malware is a big problem for cybersecurity. As bad software gets socially smarter and comes in many forms, old ways of finding and stopping it are not enough anymore. This is where models of machine learning start being useful. To summarize, a machine learning-based antivirus system can only be sustainable if it has good data, the right features, a good model, and skilled cybersecurity professionals working on it. This study focuses on Random Forest, Decision Tree, KNeighbour, AdaBoost and SGD classifiers’ various classification methodologies for detection of malware. The proposed classifiers, which may be used for future study, emerge as the best classifiers for the dataset utilized for malware detection which help in global stability.