Leveraging Machine Learning Algorithms for Threat Detection Using AI-Enhanced Cybersecurity Datasets

This study uses an AI-enhanced cybersecurity events dataset to utilize the effectiveness of leveraging machine learning algorithms Logistic Regression, Decision Tree, and Random Forest for threat detection. The dataset includes cyber incidents, which preprocess and optimize for model performance. Among the algorithms tested, Random Forest emerged as the best performer, demonstrating superior accuracy, precision, recall, and F1 score. Its ability to handle complex, high-dimensional data allows it to capture intricate patterns of cyber threats more effectively than Logistic Regression and Decision Trees. Logistic Regression, while efficient, struggles with nonlinear relationships, and Decision Tree, despite its interpretability, tends to overfit and show lower predictive accuracy. Our findings highlight Random Forest as a robust tool for enhancing threat detection in cybersecurity, advocating for using advanced ensemble methods to improve defense mechanisms against evolving cyber threats.