For high quality and trusted service selection problem, a community trust driven service selection model (CTDSS) was put forward. Based on bipartite network for service selection, the user preference similarity was formed by combining the service selection similarity with the service evaluation similarity between the cloud service consumers. Then, a clustering algorithm for trusted recommendation community was proposed. Through making use of the user community, the user's trust evaluation to an unknown service can be predicted to guide his service selection. The experimental results of the public data set and the simulated data set show that the method can effectively identify the user community based on the service preferences in cloud environment. And it can provide accurate decision-making basis for the user's service selection by using the service selection mechanism driven by community trust.