JavaScript Vulnerability Detection using Random Forest with Stratified Sampling

This KNIME workflow leverages a machine learning approach, specifically the Random Forest algorithm, to identify potential vulnerabilities in JavaScript code. It operates on a dataset ("JSVulnerabilityDataSet-1.0.csv") comprising features extracted from JavaScript code snippets, such as code complexity metrics (e.g., cyclomatic complexity, lines of code) and Halstead complexity measures. The dataset also includes a binary target variable Vuln, indicating the presence (1) or absence (0) of a vulnerability.

External resources

• Challenging Machine Learning Algorithms in Predicting Vulnerable JavaScript Functions