Social networks attract lots of new users every day and ab- sorb from them information about events and facts happen- ing in the real world. The exploitation of this information can help identifying mobility patterns that occur in an urban environment as well as produce services to take advantage of social commonalities between people. In this paper we set out to address the problem of extracting urban patterns from fragments of multiple and sparse people life traces, as they emerge from the participation to social networks. To investigate this challenging task, we analyzed 13 millions Twitter posts (3 GB) of data in New York. Then we test upon this data a probabilistic topic models approach to au- tomatically extract urban patterns from location-based so- cial network data. We nd that the extracted patterns can identify hotspots in the city, and recognize a number of ma- jor crowd behaviors that recur over time and space in the urban scenario.