Comparative Analysis of Machine Learning Classifiers for Fileless Malware Detection

The ever-expanding sophistication of cyberthreats has led to the development of fileless malware, a form of stealthy and elusive malicious software that operates solely within the memory of a computer rendering the conventional methods of detection ineffective. As a response to the significant cybersecurity risk that was just recently brought to light, this research studies the application of machine learning algorithms for the proactive identification of fileless malware. This study conducts a comprehensive evaluation of various classifiers, including Random Forest, Support Vector Classifier (SVC), K-Nearest Neighbors (KNN), Logistic Regression, Naive Bayes Classifier (NB), Gradient Boosting Classifier (GB), and Decision Trees. These classifiers are applied to analyze fileless malware and non-malware samples, with a focus on assessing their performance. In addition to that, this research investigates how well machine learning models can adapt to new fileless malware strategies when applied to real-world circumstances. These findings shed light on how effective machine learning models are in dynamically shifting threat scenarios. This research also analyzes the seamless integration of AI -driven solutions into existing security frameworks. As a result, this experimental work delivers valuable insights that can be utilized to better cybersecurity tactics in the face of a threat that is always evolving in the internet world.