International Journal of Academic Information Systems Research (IJAISR)

Title: Predicting Fire Alarms in Smoke Detection using Neural Networks

Authors: Maher Wissam Attia, Baraa Akram Abu Zaher, Nidal Hassan Nasser, Ruba Raed Al-Hour, Aya Haider Asfour, Samy S. Abu-Naser

Volume: 7

Issue: 10

Pages: 26-33

Publication Date: 2023/10/28

Abstract:
This research paper presents the development and evaluation of a neural network-based model for predicting fire alarms in smoke detection systems. Using a dataset from Kaggle containing 15 features and 3487 samples, we trained and validated a neural network with a three-layer architecture. The model achieved an accuracy of 100% and an average error of 0.0000003. Additionally, we identified the most influential features in predicting fire alarms.

Download Full Article (PDF)