CityGuard: Citywide Fire Risk Forecasting Using A Machine Learning Approach
summary
Proposes a machine learning model called NeuroFire that models internal temporal signals (historical incidents), external temporal signals (weather) and spatial dependencies using a spatial‑temporal loss, and reports that it outperforms 9 baselines on real city datasets. Implements a production system CityGuard that visualizes block‑level risk forecasts, shows visualizations of errors and case studies, and demonstrates practical deployment potential for guiding fire prevention resource placement.
AI-written summary · about
related talks
58mImproving disaster response with machine learning
46mSakshi Babbar of Amity Online at Machine Learning Developers Summit 2019
26mSridhar Telidevara: Professor, Great Lakes Institute of Management at Cypher 2017 - Part 1/2
32mPredicting Grid Faults Before They Happen: AI for Weather-Driven Grid Resilience
32mMeasuring Vegetation Health to Predict Natural Hazards
22mCan AI Support Our City Trees in Changing Climate?
09mGen AI Workflows for Climate Risk Modeling in Food Systems
37mForecasting using machine learning methods