Understanding time use via data mining: A clustering-based framework

Abstract

In this work, a data mining framework is proposed to improve the understanding of how people allocate their time. Using a multivariate approach, we performed a clustering procedure, and subsequently a regression analysis to detect which variables influence individual time use for each cluster found. Results suggest that the impact of various sociodemographic variables on sleep and work depends significantly on the characteristics of the individuals analyzed. This suggests that inquiries into time allocation and individual behavior should no longer be limited to discussions focused only on single variables. Based on our results, we recommend that researchers advance their methodological analysis towards a multifactorial approach and include clustering as a fundamental step. Proper identification of the most significant variables involved in time allocation decisions would allow researchers to better analyze and interpret their data and results.

In this work, a data mining framework is proposed to improve the understanding of how people allocate their time. Using a multivariate approach, we performed a clustering procedure, and subsequently a regression analysis to detect which variables influence individual time use for each cluster found. Results suggest that the impact of various sociodemographic variables on sleep and work depends significantly on the characteristics of the individuals analyzed. This suggests that inquiries into time allocation and individual behavior should no longer be limited to discussions focused only on single variables. Based on our results, we recommend that researchers advance their methodological analysis towards a multifactorial approach and include clustering as a fundamental step. Proper identification of the most significant variables involved in time allocation decisions would allow researchers to better analyze and interpret their data and results.