Research Article
Read the full article: Long-Term Occupancy-Estimation Algorithm using CO₂ Levels and Temporal Features in Laboratory Environments
Research Overview
This study presents a deep learning approach for predicting occupancy patterns using CO₂ sensor data and temporal features. The dataset consists of 19,189 samples collected over a year-long period (September 18, 2023 – November 21, 2024) in a university laboratory via the Smart Indoor Air Quality Monitor. A sequential neural network implemented in TensorFlow demonstrated an accuracy of 0.97 and an F1-score of 0.92. The research highlights the utility of CO₂ levels and temporal features for occupancy estimation, with advantages including low computational requirements, cost-effective sensors, an IoT-enabled interface, and scalability. Note that the study is limited to indoor environments, which may affect its generalizability.
Research Dataset
The dataset features multimodal data with CO₂ sensor readings (ranging from 0–5,000 ppm) and corresponding timestamps, providing a unique long-term record for occupancy detection research.
Open License for Research
This research dataset is provided under an open license, ensuring that it can be freely used, modified, and shared for non-commercial research purposes with proper attribution. If you use this data, please cite the associated research article accordingly. For further details, please review the Creative Commons Attribution 4.0 International License.