Table 2 Comparison with state-of-the-art approaches for occupancy estimation and \(\textrm{CO}_2\,\)sensing.

Accurate occupancy detection⁴⁰

Temperature, \(\textrm{CO}_2\,\), humidity, lighting

Linear Discriminant Analysis (LDA), Classification and Regression Trees (CART) and Random Forest (RF) models

99.5%

Occupancy prediction⁴¹

\(\textrm{CO}_2\,\), humidity, presence, etc.

Machine learning classifier algorithms, Random Forest model

at least 90%

Occupancy-count estimation using⁴²

Wi-Fi access points, \(\textrm{CO}_2\,\)sensors, PIR motion detectors, and plug and light electricity load meters

Multiple linear regression models, Artificial neural network models

80% - 83%

The detection and prediction of occupancy⁴³

Detection and recognition of occupants’ activities

Demand-driven deep learning model based on a convolutional neural network (CNN)

>80%