Data-driven chimney fire risk prediction using machine learning and point process tools

TL;DR

This paper presents a novel approach combining machine learning and statistical point process modeling to accurately predict chimney fires, enabling better prevention strategies based on data-driven insights.

Contribution

It introduces a hybrid method that uses random forests for variable selection and Poisson point process models with logistic regression for efficient, region-specific fire risk prediction.

Findings

Effective identification of key predictors using random forests.

Plausible and validated fire risk predictions on real data.

Enhanced prediction accuracy over existing methods.

Abstract

Chimney fires constitute one of the most commonly occurring fire types. Precise prediction and prompt prevention are crucial in reducing the harm they cause. In this paper, we develop a combined machine learning and statistical modeling process to predict chimney fires. Firstly, we use random forests and permutation importance techniques to identify the most informative explanatory variables. Secondly, we design a Poisson point process model and apply associated logistic regression estimation to estimate the parameters. Moreover, we validate the Poisson model assumption using second-order summary statistics and residuals. We implement the modeling process on data collected by the Twente Fire Brigade and obtain plausible predictions. Compared to similar studies, our approach has two advantages: i) with random forests, we can select explanatory variables non-parametrically considering variable dependence; ii) using logistic regression estimation, we can fit the statistical model efficiently by tuning it to focus on important regions and times of the fire data.