Fatal structure fire classification from building fire data using machine learning

Abstract

Purpose

This study aims to develop a machine learning model to detect structure fire fatalities using a dataset comprising 11,341 cases from 2011 to 2019.

Design/methodology/approach

Exploratory data analysis (EDA) was conducted prior to modelling, in which ten machine learning models were experimented with.

Findings

The main fatal structure fire risk factors were fires originating from bedrooms, living areas and the cooking/dining areas. The highest fatality rate (20.69%) was reported for fires ignited due to bedding (23.43%), despite a low fire incident rate (3.50%). Using 21 structure fire features, Random Forest (RF) yielded the best detection performance with 86% accuracy, followed by Decision Tree (DT) with bagging (accuracy = 84.7%).

Research limitations/practical implications

Limitations of the study are pertaining to data quality and grouping of categories in the data pre-processing stage, which could affect the performance of the models.

Originality/value

The study is the first of its kind to manipulate risk factors to detect fatal structure classification, particularly focussing on structure fire fatalities. Most of the previous studies examined the importance of fire risk factors and their relationship to the fire risk level.

Citation

Balakrishnan, V., Mohammed Hashim, A.N., Lee, V.C., Lee, V.H. and Lee, Y.Q. (2023), "Fatal structure fire classification from building fire data using machine learning", International Journal of Intelligent Computing and Cybernetics, Vol. ahead-of-print No. ahead-of-print. https://doi.org/10.1108/IJICC-07-2023-0167