Towards Better Understanding of Deep Learning Model for Time Series Forecasting in Fire Research

Published

July 29, 2024

Author(s)

Linhao Fan, Qi Tong, Hongqiang FANG, Wei Zhong, Wai Cheong Tam, Tianshui Liang

Abstract

Deep learning model has been a viable approach to forecast critical events in fire development. However, prior to its implementation in real-life firefighting, it is imperative to further understand the black box and assess its rationale. In this paper, an interpretability analysis framework was proposed to reliably enhance the transparency of deep learning models in time series. The framework was applied to a flashover forecasting model as a case study, including employing the interpretability method to obtain attributions and adapting the evaluation metrics to validate the method's effectiveness and determine its optimal parameter setting for the model. Results show that the proposed interpretability method, named DeepLIFT, can provide precise attributions to the model inputs in both temporal and spatial domains. Based on the quantitative analysis, suitable parameters were found and the relevance of the attribution results to the model decision was validated, which means the attribution results are reliable to be utilized to interpret the model. It is believed this work would contribute to bringing trustworthy deep learning models for fire research.

Citation

Fire Technology

Pub Type

Journals

Download Paper

Local Download

Keywords

black-box model, flashover, model interpretability, trustworthy AI, time series classification, smart firefighting

Citation

Fan, L. , Tong, Q. , FANG, H. , Zhong, W. , Tam, W. and Liang, T. (2024), Towards Better Understanding of Deep Learning Model for Time Series Forecasting in Fire Research, Fire Technology, [online], https://tsapps.nist.gov/publication/get_pdf.cfm?pub_id=957125 (Accessed December 27, 2024)