An Incremental Majority Voting Approach for Intrusion Detection System Based on Machine Learning

With the exponential growth of digitalization and data volumes, the cybersecurity threat landscape hasbecome increasingly complex, amplifying the need for robust intrusion detection systems (IDS). TraditionalIDS approaches often struggle with static architectures, requiring costly and frequent retraining to keep upwith evolving threats. This study introduces an incremental, majority-voting IDS system that leveragesmachine learning to adapt to continuous network traffic streams without the need for extensive retraining.By integrating multiple machine learning algorithms—K-Nearest Neighbors (KNN), Logistic Regression,Bernoulli Naive Bayes, and Decision Tree classifiers—the system employs a collective decision-makingapproach to enhance detection accuracy and minimize false alarms in real-time. Results indicate that thismulti-algorithm IDS framework offers substantial improvements in adaptability, performance, andresilience against intrusions, especially within real-world, imbalanced data scenarios.