Models versus Datasets: Reducing Bias through Building a Comprehensive IDS Benchmark

Abstract

Today, deep learning approaches are widely used to build Intrusion Detection Systems for securing IoT environments. However, the models’ hidden and complex nature raises various concerns, such as trusting the model output and understanding why the model made certain decisions. Researchers generally publish their proposed model’s settings and performance results based on a specific dataset and a classification model but do not report the proposed model’s output and findings. Similarly, many researchers suggest an IDS solution by focusing only on a single benchmark dataset and classifier. Such solutions are prone to generating inaccurate and biased results. This paper overcomes these limitations in previous work by analyzing various benchmark datasets and various individual and hybrid deep learning classifiers towards finding the best IDS solution for IoT that is efficient, lightweight, and comprehensive in detecting network anomalies. We also showed the model’s localized predictions and analyzed the top contributing features impacting the global performance of deep learning models. This paper aims to extract the aggregate knowledge from various datasets and classifiers and analyze the commonalities to avoid any possible bias in results and increase the trust and transparency of deep learning models. We believe this paper’s findings will help future researchers build a comprehensive IDS based on well-performing classifiers and utilize the aggregated knowledge and the minimum set of significantly contributing features.