Reproduce “Optimizing Fire Safety: Reducing False Alarms Using Advanced Machine Learning Techniques” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Muhammad Hassan Jamal, Abdulwahab Alazeb, Shahid Allah Bakhsh, Wadii Boulila, Syed Aziz Shah, Aizaz Ahmad Khattak, Muhammad Shahbaz Khan
This research develops a weighted ensemble machine learning model that addresses the issue of false fire alarms by distinguishing between genuine fire conditions and non-fire situations. The model achieves 99.98% accuracy with only 1 false positive and 2 false negatives out of 17,903 test cases on a public smoke detection dataset, significantly enhancing fire safety system reliability.
Key insights
- Class imbalance in fire detection datasets can be effectively addressed using techniques like SMOTE to prevent model bias.
- A density-based weighted ensemble approach, combining KNN's local pattern recognition with XGBoost's global pattern handling, can effectively leverage the strengths of different algorithms for robust fire detection.
- Environmental factors such as humidity and pressure are strong predictors in distinguishing genuine fire conditions from normal sensor readings.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.