ata mining techniques combined with space-time cubes empower the analysis of multidimensional data. A study area in which these advanced analysis techniques can be applied pre-eminently is urban mobility, where investigation of non-motorized mobility patterns is a main priority for several cities around the world. The presented work aimed to extract spatio-temporal patterns from a human movement database containing volunteer-generated cycling data in Cuenca (Ecuador) with the objective to detect places and times where strategies can be applied that promote urban cycling. The methodology takes advantage of the capabilities of the space-time pattern mining toolbox in ArcGIS. The results demonstrate the viability of the proposed methodology for the characterization of non-motorized mobility patterns and its potential for analyzing other mobility datasets.