Accurately ranking images and multimedia objects is of paramout relevance in many retrieval and learning tasks. Manifold learning methods have been investigated for ranking mainly due to its capacity of taking into account the intrinsic global manifold structure. In this paper, a novel manifold ranking algorithm is proposed based on hypergraphs for unsupervised multimedia retrieval tasks. Different from traditional graphbased approaches, which represents only pairwise relationships, hypergraphs are capable of modeling similarity relationships among set of objects. The proposed approach uses the hyperedges for constructing a contextual representation of data samples, and exploits the encoded information for deriving a more effective similarity function. An extensive experimental evaluation was conducted on 10 public datasets, including diverse retrieval scenarios and multimedia content. Experimental results demonstrate that high effectiveness gains can be obtained in comparison with state-of-the-art methods.