Catalog Summary
"Real-time data collected from IoT sensors to detect smoke and fire events. Features include temperature, humidity, TVOC, and eCO2 levels for reliable alarm systems."
Catalog Notes
Dataset Context
This dataset is crucial for developing AI-based fire alarm systems. Traditional smoke detectors can be prone to false alarms; this IoT dataset allows for multi-sensor fusion to improve accuracy.
Technical Details
- Device: Photoelectric Smoke Detector combined with IoT sensors.
- Target Variable: 'Fire Alarm' (1 = Fire, 0 = No Fire).
- Key Features:
- Temperature (C) & Humidity (%)
- TVOC (Total Volatile Organic Compounds) [ppb]
- eCO2 (CO2 equivalent) [ppm]
- Raw Sensor Data (Raw H2, Raw Ethanol)
Applications
Perfect for training Binary Classification models (Logistic Regression, Random Forest) to predict fire incidents with high precision.
Data Preview
| UTC | Temperature[C] | Humidity[%] | TVOC[ppb] | eCO2[ppm] | Fire_Alarm |
|---|---|---|---|---|---|
| 1654733331 | 20.0 | 57.36 | 0 | 400 | 0 |
| 1654733332 | 20.015 | 56.67 | 0 | 400 | 0 |
| 1654733333 | 20.029 | 55.96 | 0 | 400 | 0 |
Showing first few rows for preview
Guides and analyses for this dataset
Read the linked methodology, interpretation, and reproducibility guidance before analysing or reusing this dataset.
Provided Citation
Deep Contractor. (2022). Smoke Detection Dataset. Kaggle. Retrieved from https://www.kaggle.com/datasets/deepcontractor/smoke-detection-dataset
This citation is displayed as supplied. Automatic style conversion is disabled because structured citation metadata is not recorded.
Source metadata: Kaggle (2026)
Indexed by IoTDataset.com on Jan 13, 2026
Review the Source Record
Confirm the licence, version, access conditions, file format, and provenance at the source before use.