Urban regional risk is a complex nonlinear problem that encounters insufficient information, randomness, and uncertainty. To accurately assess the overall urban risk, a regional risk assessment model for urban public safety was proposed by using the information diffusion theory. The entropy theory was employed to optimize the information diffusion model to reduce the uncertainty. A framework of urban regional risk assessment model based on information diffusion and entropy was constructed. Finally, a case study of Hangzhou city in China was presented to demonstrate the performance of the proposed method. Results showed that the proposed method could successfully estimate the urban regional risk of Hangzhou city. The risk levels and probabilities of different hazard indicators were basically consistent with reality. The hazards with respect to industrial and mining accidents and road traffic accidents were extremely serious. More than 80 deaths from industrial and mining accidents would occur almost every 3 years, and more than 400 deaths of RTA would occur almost every 2.6 years. Moreover, centralized intervals of the risk level associated with five hazards were found, where urban risks were more likely to happen and had higher vulnerability. It could provide guidance for the government’s urban safety management and policy-making.