Although automatically collected human travel records can accurately capture the time and location of human movements, they do not directly explain the hidden semantic structures behind the data, e.g., activity types and/or travel purposes. Probabilistic topic models, which are widely used in natural language processing for document classification, can be used to uncover semantic human mobility patterns from large amount of spatiotemporal data in an unsupervised manner. In this case, the trips of an individual user are treated as words in a document, and each topic is a distribution over space and time that corresponds to some activity. Specifically, a classical topic model, Latent Dirichlet Allocation (LDA), is extended to incorporate multiple heterogeneous trip attributes—the destination, time of arrival, day of week, and duration of stay. The proposed methodology is demonstrated using pseudonymized transit smart card data from London, U.K. The discovered topics reveal diverse spatiotemporal patterns, and are mostly interpretable. It is relatively easy to identify work/school, home, and night-life activities from these topics. The results can be used to infer the latent activity patterns behind the physical human movements directly observed in mobility datasets.