Robustness Analysis of Machine Learning Models for IoT Intrusion Detection Under Data Poisoning Attacks

2026-04-15

AI Summary

This study addresses the severe degradation in reliability of IoT intrusion detection systems under data poisoning attacks. It systematically evaluates the robustness of four mainstream classifiers—Random Forest, Gradient Boosting Machines, Logistic Regression, and Deep Neural Networks—against label flipping and outlier injection attacks, analyzing performance deterioration mechanisms and decision boundary perturbations using three real-world IoT datasets. The work provides the first quantitative assessment of how different poisoning strategies impact IoT detection models, revealing that ensemble methods exhibit strong robustness, whereas Logistic Regression and Deep Neural Networks suffer accuracy drops of up to 40%. Building on these findings, the paper proposes integrating adversarially robust training, anomaly monitoring, and feature validation into practical deployment and compliance frameworks to enhance system security.

Technology Category

• Application Domains: Internet of Things, Sensor Networks & Smart Cities
• Machine Learning: Adversarial Learning & Robustness
• Data Mining & Knowledge Management: Anomaly/Outlier Detection

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Web Mining and Content Analysis: Web data provenance, reliability, and authenticity

Abstract

Ensuring the reliability of machine learning-based intrusion detection systems remains a critical challenge in Internet of Things (IoT) environments, particularly as data poisoning attacks increasingly threaten the integrity of model training pipelines. This study evaluates the susceptibility of four widely used classifiers, Random Forest, Gradient Boosting Machine, Logistic Regression, and Deep Neural Network models, against multiple poisoning strategies using three real-world IoT datasets. Results show that while ensemble-based models exhibit comparatively stable performance, Logistic Regression and Deep Neural Networks suffer degradation of up to 40% under label manipulation and outlier-based attacks. Such disruptions significantly distort decision boundaries, reduce detection fidelity, and undermine deployment readiness. The findings highlight the need for adversarially robust training, continuous anomaly monitoring, and feature-level validation within operational Network Intrusion Detection Systems. The study also emphasizes the importance of integrating resilience testing into regulatory and compliance frameworks for AI-driven IoT security. Overall, this work provides an empirical foundation for developing more resilient intrusion detection pipelines and informs future research on adaptive, attack-aware models capable of maintaining reliability under adversarial IoT conditions.

Problem

• IoT intrusion detection
• data poisoning attacks
• machine learning robustness
• adversarial threats
• model reliability

Innovation

• data poisoning attacks
• IoT intrusion detection
• adversarial robustness
• machine learning reliability
• resilience testing