To enable Industry 4.0 successfully, there is a need to build a resilient automation system that can quickly recover after having been attacked or robustly sustain continued operations while being threatened, enable an automated monitoring evolution via various sensor channels in real time, and use advanced machine learning and data analytics to formulate strategies to mitigate and eliminate faults, threats, and malicious attacks. It is envisioned that if we can develop an intelligent model that (a) represents a meaningful, realistic environment and complex entity containing manufacturing Internet of Things interdependent and independent properties that are stepping-stones of the cyber kill chain or precursors of the onset of cyberattacks; (b) can learn and predict potential errors and formulate offense/defense strategies and healing solutions; (c) can enable cognitive ability and human-in-the-loop analytics in real time; and (d) can facilitate system behavior changes to disrupt the attack cascade, then the hosting system can learn how to neutralize threats and attacks and self-repair infected or damaged links autonomously. In this article, our preliminary work presents a visual analytics framework and technique for situational awareness, including autonomously monitoring, diagnosing, and prognosticating the state of cyber-physical systems. Our approach, presented in this article, relies on visual characterizations of multivariate time series and real-time predictive analytics to highlight potential faults, threats, and malicious attacks. To validate the usefulness of our approach, we demonstrate the developed technique using various aviation datasets obtained from the Prognostics Center of Excellence at the National Aeronautics and Space Administration Ames.
Figure 1: Visualizing the prevalence of HIV (male vs. female) using Outliagnostics: (a) The control panel, (b) the lensing area, (c) the scatterplot series, (d) the top countries clouds, (e) the customized outlying boxplots, and (f) the country outlying profiles.
ABSTRACTThis paper presents an approach to analyzing two-dimensional temporal datasets focusing on identifying observations that are significant in calculating the outliers of a scatterplot. We also propose a prototype, called Outliagnostics, to guide users when interactively exploring abnormalities in large time series. Instead of focusing on detecting outliers at each time point, we monitor and display the discrepant temporal signatures of each data entry concerning the overall distributions. Our prototype is designed to handle these tasks in parallel to improve performance. To highlight the benefits and performance of our approach, we illustrate and validate the use of Outliagnostics on real-world datasets of various sizes in different parallelism configurations. This work also discusses how to extend these ideas to handle time series with a higher number of dimensions and provides a prototype for this type of datasets.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.