• Priya N

The sheer quantity of data and files that need to be analysed for potential threats is the greatest challenge for malware researchers. On a daily basis, researchers find and analyse a significant number of new malwares in order to extract common characteristics. Therefore, it is essential for the investigation of malware characteristics to have a scheme that can ensure and improve the efficacy and accuracy of categorization. This paper proposes a very efficient automated classification method that combines multi-feature selection and machine learning-based fusion. The system demonstrates great performance. Based on our evaluations, it exhibits superior performance and functionality compared to single-featured devices. The proliferation of malware poses a substantial threat as its use continues to expand. Manual heuristic malware assessment is no longer deemed efficient owing to the fast propagation of malware. Therefore, the use of machine learning algorithms to automatically identify and analyse malicious software behaviour is considered a very effective option. Behaviour reports will definitely be generated after the analysis of each malware's behaviour in a simulated environment. It is necessary to pre-process such data by converting them into sparse trajectory models before using machine learning techniques, specifically for classification purposes. The research used Support Vector Machines (SVM) and Random Forests as classifiers. Ultimately, using autonomous behaviour and machine learning techniques may effectively and efficiently detect and classify malware, as shown by this proof of concept. Due to the substantial negative impact that many antivirus programmes have on the user's system and their occasional inability to detect new and hazardous threats, lightweight antivirus programmes are less efficient in discovering and preventing malware. Our approach uses a cloud-based malware classification algorithm, with the main focus being on safeguarding data while minimising any noticeable impact on the user's system. Consequently, we can assess and ascertain if any dubious file is malicious or not without the need to install additional third-party software on user PCs

Malware Detection, Machine Learning, Cyber Security, Network Security

"MALWARE DETECTION USING MACHINE LEARNING", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.11, Issue 4, page no.e684-e692, April-2024, Available :http://www.jetir.org/papers/JETIR2404474.pdf

"MALWARE DETECTION USING MACHINE LEARNING", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.11, Issue 4, page no. ppe684-e692, April-2024, Available at : http://www.jetir.org/papers/JETIR2404474.pdf