2019/11/11 by Albert M. Lund, Lund, Albert M, Ramkiran Gouripeddi +3
Computer Science · Social Sciences · #Context-Aware Activity Recognition Systems #FOS: Computer and information sciences #Human Mobility and Location-Based Analysis #Social and Information Networks (cs.SI) #Urban Transport and Accessibility
paper · pdf · doi:10.48550/arxiv.1911.05476
openalex publication_date 2019/11/11 · openalex created_date 2022/07/26 · openalex updated_date 2026/07/28
Human activity encompasses a series of complex spatiotemporal processes that\nare difficult to model, but represents an essential component of human exposure\nassessment. A significant empirical data source like the American Time Use\nSurvey (ATUS) can be leveraged to model human activity, but tractable models\nrequire a better stratification of activity data to inform about different, but\nclassifiable groups of individuals that exhibit similar activities and mobility\npatterns. We have developed a simple unsupervised classification and sequence\ngeneration method from existing machine learning algorithms that is capable of\ngenerating coherent and stochastic sequences of activity from the data in the\nATUS. This classification, when combined with any spatiotemporal exposure\nprofile, allows the development of stochastic models of exposure patterns for\ngroups of individuals exhibiting similar activity behaviors.\n