The Personalized Thermal Comfort Prediction Using an MH-LSTM Neural Network Method

Table 1

Previous studies.

Author	Research scope	Target	Highlighted factors						Research methodology
			F1	F2	F3	F4	F5	F6
Yao et al. [24]	Self-regulatory actions	Psychological factor	—	—	○	—	—	—	Survey and monitoring
Jowkar et al. [26]		Thermal exposure	—	—	○	—	—	○	Field study, classification, and data analysis
Xiong et al. [27]	Physiological parameter	Investigated gender differences	○	—	○	○	—	—	Experiment
Choi and Yeom [28]		Local skin temperatures	○	—	○	○	—	○	Monitoring
Chaudhuri et al. [29]	Machine learning	Predicted thermal state (PTS)	—	○	○	—	—	○	Machine learning
Cosma et al. [30]		Individual thermal preference model	○	○	○	○	—	○	Experiment
Katic et al. [31]		Artificial neural network algorithm	—	○	—	—	—	○	Data analysis

Note. F1 = skin temperature; F2 = artificial intelligence approaches; F3 = gender; F4 = sensor device; F5 = individual characteristics; F6 = prediction of thermal comfort.