Prediction of Occupant Behavior toward Natural Ventilation in Japanese Dwellings: Machine Learning Models and Feature Selection

Occupant behavior based on natural ventilation has a significant impact building energy consumption. It is important for the quantification of occupant-behavior models to select observed variables, i.e., features that affect state window opening and closing, consider machine learning are effective in predicting this state. In study, thermal comfort was investigated, data were analyzed 30 houses Gifu, Japan. Among selected models, logistic regression deep neural network produced consistently excellent results. The accuracy prediction open closed windows differed among factors influencing window-opening behaviors occupants from those their window-closing behavior. selection features, analysis using indices representative room cooling showed results, indicating which have conflicting relationships with ventilation, useful improving prediction. present study indicates designers should incorporate occupant into designs.