This paper introduces a framework for inferring activity travel given nearby land use information that can be obtained from a location-based social network (LBSN) such as Foursquare. The first component of the framework implements a generic method for acquiring land use data from LBSN services, which is a prerequisite for the inference algorithm. Three inference algorithms are suggested, and situations in which each algorithm might be a better fit are discussed. Finally, a case study is presented for activity inference applied to a data set collected in the greater Toronto and Hamilton area, Ontario, Canada, during the fall of 2012. Results are encouraging and suggest that it is possible to infer daily activity travel automatically; this possibility could significantly reduce the burdens of personal travel surveys and allow for collection of long-period travel diary data that is not easily achievable with traditional survey methods.