Table 1 Classification performance of the Attention-LSTM model thermal comfort classes.

From: Digital twin based deep learning framework for personalized thermal comfort prediction and energy efficient operation in smart buildings

Class

Precision

Recall

F1-Score

Neutral (N)

0.84

0.99

0.91

Uncomfortably Cold (UC)

0.92

0.36

0.52

Uncomfortably Warm (UW)

0.50

0.02

0.05

Overall Accuracy

0.838 (83.8%)

Back to article page