This paper made a short review of Cloud Computing and Big Data, and discussed the portability of general data mining algorithms to Cloud Computing platform. It revealed the Cloud Computing platform based on Map-Reduce cannot solve all the Big Data and data mining problems. Transplanting the general data mining algorithms to the real-time Cloud Computing platform will be one of the research focuses in Cloud Computing and Big Data.

At present, the main conventional method to study pollution diffusion is to establish a mathematical model to analyse the relationship between pollution sources. The lack of analysis on the characteristics and attributes of pollution sources leads to unsatisfactory identification results. To solve this problem, a pollution diffusion prediction method based on pollution atlas was proposed. The pollution knowledge graph was constructed to describe the relationship between pollution source attributes and the interaction and influence between pollution source and region. The diffusion characteristics of pollution were extracted from the atlas, and DBSCAN was used to achieve clustering and deep neural network model was built to predict the pollution degree at different depths. The model is verified in a typical contaminated site. The experimental results show that the proposed method is effective and has good recognition effect and convergence speed.