Data Science for High-Stakes Decisions – Crime Analytics, Environmental Illegality, and Explainability
Time: Thu 2026-05-28 13.00 - 14.00
Location: Digital Futures hub, Osquars Backe 5, floor 2 at KTH main campus OR Zoom
Video link: https://kth-se.zoom.us/j/69560887455
Speaker: Luis Gustavo Nonato, University of São Paulo, Brazil
Abstract: This presentation provides an overview of my research group’s efforts to develop mathematical and computational frameworks for analyzing high-stakes systems. Our research is organized around three interconnected pillars.
-
Crime Analytics focuses on modeling spatio-temporal data to support the understanding of crime dynamics and enhance public safety.
-
Timber Trade Network Analysis applies network science methods to characterize trade structures and to distinguish legal from illegal flows, contributing to environmental monitoring and governance.
-
Explainable Artificial Intelligence (XAI) addresses the need for transparency, interpretability, and trust in machine learning models deployed in sensitive societal and environmental contexts.
The overarching goal of this presentation is to provide a panoramic view of these research directions to foster opportunities for future interdisciplinary collaborations.