A Data Mining Approach to Analyze Occupant Behavior Motivation. (2017)

• Record Type: Journal Article
• Authors: Ren, Xinyuyang; Zhao, Yang; Zeiler, Wim; Boxem, Gert; Li, Tingting
• Is Part Of: Procedia engineering. Volume 205(2017)
• Page Start: 2442
• Page End: 2448
• Publication Date: 2017
• Subjects: Data mining -- Occupant behavior -- Motivation pattern; Engineering -- Congresses; Engineering -- Periodicals; Engineering; Conference proceedings; Periodicals; 620.005
• DOI: 10.1016/j.proeng.2017.09.971
• Languages: English
• ISSNs: 1877-7058

Abstract

Abstract: Occupants' behavior could bring significant impact on the performance of built environment. Methods of analyzing people's behavior have not been adequately developed. The traditional methods such as survey or interview are not efficient. This study proposed a data-driven method to analyze the occupants' behavior, supported by a specific case of analyzing people's adjustment to ventilation system in a Dutch community. In the individual level, to analyze the motivation of a single person, a logistic regression based approach was proposed to classify occupants' behavior of increasing/decreasing the ventilation flowrate and then reveal the motivations behind. In the community level, the behavior motivations derived from different occupants were compared. Three motivational behavior patterns, namely the environment-driven type, the time-driven type and the mixed-type were summarized. The proposed mining method is useful to discover and develop occupant behavior models.