My research is much interdisciplinary or cross-disciplinary in nature. I looked at urban structure and dynamics using massive geographic information and based on complexity modeling tools such as agent-based modeling and complex networks. In particular, I developed a topological analysis of urban street networks, and found some interesting scaling pattern, as demonstrated in many other complex networks including biological, information and technological. My recent work based on 40 US cities illustrated a universal pattern that can be simply yet elegantly characterized by the 80/20 principle, i.e. 80 percent of streets are less connected (less than 4 other streets, and 4 being an average), while 20 percent of streets are well connected (above the average); out of the 20 percent, there is less 1 percent of streets that are extremely well connected. More recently, we have explored such street hierarchies based on massive GPS tracking log, and proved that streets are indeed hierarchically organized in the sense that a minority of streets accounts for a majority of traffic flow. Further more, human movement pattern at a collective level is mainly shaped by the underlying street structure, and has little to do with human moving behavior (http://fromto.hig.se/~bjg/movingbehavior/). All the research leads to the belief that cities, although artifacts in nature, are self-organized in the same way as biological entities.