Challenging Machine Learning Algorithms in Predicting Vulnerable JavaScript Functions

TL;DR

This study evaluates various machine learning algorithms, including deep learning, for predicting security vulnerabilities in JavaScript functions, using a new dataset and static code metrics, with KNN showing the best performance.

Contribution

It introduces a new dataset for vulnerability prediction in JavaScript and compares multiple ML algorithms, highlighting the effectiveness of KNN and the impact of re-sampling strategies.

Findings

KNN achieved an F-measure of 0.76 in vulnerability prediction.

Deep learning and SVM classifiers performed competitively with F-measures over 0.70.

Re-sampling strategies affected the balance between precision and recall.

Abstract

The rapid rise of cyber-crime activities and the growing number of devices threatened by them place software security issues in the spotlight. As around 90% of all attacks exploit known types of security issues, finding vulnerable components and applying existing mitigation techniques is a viable practical approach for fighting against cyber-crime. In this paper, we investigate how the state-of-the-art machine learning techniques, including a popular deep learning algorithm, perform in predicting functions with possible security vulnerabilities in JavaScript programs. We applied 8 machine learning algorithms to build prediction models using a new dataset constructed for this research from the vulnerability information in public databases of the Node Security Project and the Snyk platform, and code fixing patches from GitHub. We used static source code metrics as predictors and an extensive grid-search algorithm to find the best performing models. We also examined the effect of various re-sampling strategies to handle the imbalanced nature of the dataset. The best performing algorithm was KNN, which created a model for the prediction of vulnerable functions with an F-measure of 0.76 (0.91 precision and 0.66 recall). Moreover, deep learning, tree and forest based classifiers, and SVM were competitive with F-measures over 0.70. Although the F-measures did not vary significantly with the re-sampling strategies, the distribution of precision and recall did change. No re-sampling seemed to produce models preferring high precision, while re-sampling strategies balanced the IR measures.