Bridging the Gap: A Comparative Analysis of ICS and IT Datasets for IDS Evaluation

This analysis addresses the critical knowledge gap in analyzing, preprocessing, and applying machine learning algorithms to Operational Technology data by comparing it to the ease of applying the same process to Informational Technology data in the domain of Intrusion Detection. The challenge lies in processing multivariate time series data like Operational Technology data, which requires both statistical analysis and modeling. Common tests, such as stationarity and causality checks, are computationally expensive and difficult to implement due to the complex nature of the data. Unlike Operational Technology data, Information Technology data we used is not a time series, making it easier to analyze and preprocess. This difference in structure creates significant barriers to applying the same workflows effectively across both domains. Thus, this research provides a comprehensive explanation of the limitations encountered when applying various methods to Operational Technology data compared to Information Technology data by carefully comparing each case.