Aiming at the deficiencies of existing context-aware recommendation algorithms, this paper proposes a context-aware recommendation algorithm based on fuzzy C-means clustering. Firstly, the fuzzy C-means clustering algorithm is used to perform clustering of historical contextual information and produce clusters and membership matrix. Then contextual information for the active user is matched with the cluster of historical contextual information, and non-membership data, which accord with the condition, are mapped into membership data by using membership degree of clustering as a mapping coefficient. Finally, we choose ratings of membership users, which conform to the active contextual information, to generate recommendation for the user. Compared with the existing algorithms, the proposed algorithm can not only solve the problem of inaccuracy recommendation due to the change of user context, but also overcome traditional hard clustering by using fuzzy C-means clustering algorithm. Moreover, the problem of data sparseness caused by clustering is solved by using membership mapping function. The experiments are conducted on the MovieLens dataset and the effectiveness of the proposed algorithm is verified.