Predicting the thermal sensations of building occupants is challenging, but useful for indoor environment conditioning. In this study, a data-driven thermal sensation prediction model was developed using three quality-controlled thermal comfort databases. Different machine-learning algorithms were compared in terms of prediction accuracy and rationality. The model was further improved by adding categorical inputs, and building submodels and general models for different contexts. A comprehensive data-driven thermal sensation prediction model was established. The results indicate that the multilayer perceptron (MLP) algorithm achieves higher prediction accuracy and more rational results than the other four algorithms in this specific case. Labeling AC and NV scenarios, climate zones, and cooling and heating seasons can improve model performance. Establishing submodels for specific scenarios can result in better thermal sensation vote (TSV) predictions than using general models with or without labels. With 11 submodels corresponding to 11 scenarios, and three general models without labels, the final TSV prediction model achieved higher prediction accuracy, with 64.7%–90.7% fewer prediction errors (reducing SSE by 3.2–4.9) than the predicted mean vote (PMV). Possible applications of the new model are discussed. The findings of this study can help in development of simple, accurate, and rational thermal sensation prediction tools.

When studying the thermal adaptation of building occupants, understanding the effects of different thermal experiences on adaptation is necessary, particularly for moderate and severe heat exposure. However, this area has seen limited research. Further, skin temperature, a common parameter for quantifying thermal sensation, may insufficiently reflect the automatic thermoregulation of the human body. This study investigates the effects of long-term heat exposure on the human body using multiple physiological and subjective indexes. Two heat exposure experiments were conducted on healthy male participants from northern and southern China. Participant responses, including skin temperature, heart rate, heart rate variability, blood volume pulse (BVP), subjective thermal comfort, thermal sensation, thermal acceptability, and normalized high and low frequency values were collected and compared. The results indicated that the subjective responses of northern and southern participants were not significantly different; however, the subjective physiological symptoms and self-reported discomfort of the latter were less than those of the former, indicating that the southern participants had superior heat tolerance. Additionally, the physiological responses of all the participants were largely similar. However, southern participants showed slightly higher normalized high frequency and BVP values, indicating that they have more active vagus nerves and better vasodilation. They also showed a wider acceptable temperature range and better acclimation to heat exposure. Notably, the mean skin temperature could not effectively predict thermal sensation during heat exposure; this was more accurately achieved using the rate of change of skin temperature. These findings suggest that long-term thermal experiences can affect building occupants’ thermal adaptability.