Ramsey, A., Kale, A., Kassa, Y., Gandhi, R., & Ricks, B. (2023). Toward Interactive Visualizations for Explaining Machine Learning Models. In Jaziar Radianti, Ioannis Dokas, Nicolas Lalone, & Deepak Khazanchi (Eds.), Proceedings of the 20th International ISCRAM Conference (pp. 837–852). Omaha, USA: University of Nebraska at Omaha.
Abstract: Researchers and end users generally demand more trust and transparency from Machine learning (ML) models due to the complexity of their learned rule spaces. The field of eXplainable Artificial Intelligence (XAI) seeks to rectify this problem by developing methods of explaining ML models and the attributes used in making inferences. In the area of structural health monitoring of bridges, machine learning can offer insight into the relation between a bridge’s conditions and its environment over time. In this paper, we describe three visualization techniques that explain decision tree (DT) ML models that identify which features of a bridge make it more likely to receive repairs. Each of these visualizations enable interpretation, exploration, and clarification of complex DT models. We outline the development of these visualizations, along with their validity by experts in AI and in bridge design and engineering. This work has inherent benefits in the field of XAI as a direction for future research and as a tool for interactive visual explanation of ML models.
|
|
Abdelgawad, A. A. (2023). An Updated System Dynamics Model for Analysing the Cascading Effects of Critical Infrastructure Failures. In Jaziar Radianti, Ioannis Dokas, Nicolas Lalone, & Deepak Khazanchi (Eds.), Proceedings of the 20th International ISCRAM Conference (pp. 595–608). Omaha, USA: University of Nebraska at Omaha.
Abstract: Aiming at examining the cascading effects of the failure of Critical Infrastructure (CI), this work-in-progress research introduces an improved System Dynamics model. We represent an improvement over the previous models aimed at studying CIs interdependencies and their cascading effects. Our model builds on earlier models and corrects their flaws. In addition to introducing structural enhancements, the improvements include using unpublished data, a fresh look at a previously collected dataset and employing a new data processing to address and resolve some longstanding issues. The dataset was fed to an optimisation model to produce a new dataset used in our model. The structure of our SD model, its dataset and the data processing techniques we employed to create this dataset are all described in the study. Although the model has passed the fundamental validation criteria, more validation testing and scenario exploration are yet to be conducted.
|
|
Hager, F., Reuter-Oppermann, M., Müller, T., & Ottenburger, S. (2023). Towards the Design of a Simulation-based Decision Support System for Mass-Casualty Incidents. In Jaziar Radianti, Ioannis Dokas, Nicolas Lalone, & Deepak Khazanchi (Eds.), Proceedings of the 20th International ISCRAM Conference (pp. 565–574). Omaha, USA: University of Nebraska at Omaha.
Abstract: In case of a mass-casualty incident, e.g. due to a disaster, a high number of patients need medical care within a short time frame and often, a significant percentage must be transported to a hospital or another suitable care facility. Then, different mass transportation modes (e.g., busses, ships or trains) may be used to quickly transport patients to available medical treatment centres outside of the disaster area. Within the SimPaTrans project, we develop a simulation-based decision support system for locating, sizing and analysing different modes of transport in order to prepare for mass-casualty incidents in Germany. In this paper, we present the outline of the tool as well as a first optimisation use case for transportation patients within the city of Karlsruhe, Germany
|
|
Wang, D., & Kogan, M. (2023). Resonance+: Augmenting Collective Attention to Find Information on Public Cognition and Perception of Risk. In Jaziar Radianti, Ioannis Dokas, Nicolas Lalone, & Deepak Khazanchi (Eds.), Proceedings of the 20th International ISCRAM Conference (pp. 487–500). Omaha, USA: University of Nebraska at Omaha.
Abstract: Microblogging platforms have been increasingly used by the public and crisis managers in crisis. The increasing volume of data has made such platforms more difficult for officials to find on-the-ground information and understand the public’s perception of the evolving risks. The crisis informatics literature has proposed various technological solutions to find relevant information from social media. However, the cognitive processes of the affected population and their subsequent responses, such as perceptions, emotional and behavioral responses, are still under-examined at scale. Yet, such information is important for gauging public perception of risks, an important task for PIOs and emergency managers. In this work, we leverage the noise-cutting power of collective attention and take cues from the Protective Action Decision Model, to propose a method that estimates shifts in collective attention with a special focus on the cognitive processes of those affected and their subsequent responses.
|
|
Benaben, F., Fertier, A., Cerabona, T., Moradkhani, N., Lauras, M., & Montreuil, B. (2023). Decision Support in uncertain contexts: Physics of Decision and Virtual Reality. In Jaziar Radianti, Ioannis Dokas, Nicolas Lalone, & Deepak Khazanchi (Eds.), Proceedings of the 20th International ISCRAM Conference (pp. 54–66). Omaha, USA: University of Nebraska at Omaha.
Abstract: Virtual Reality (VR) is often used for its ability to mimic reality. However, VR can also be used for its ability to escape reality. In that case, on the one hand VR provides a visualization environment where the user’s senses are still in a familiar context (one can see if something is in front, behind, up, down, far or close), yet on the other hand, VR allows to escape the usual limits of reality by providing a way to turn abstract concepts into concrete and interactive objects. In this paper, the dynamic management of a complex industrial system (a supply chain) is enabled in a VR prototypical environment, through the management of a physical trajectory that can be deflected by the impact of any potentialities such as risks or opportunities, seen as physical objects in the performance space.
|
|