Abstract: Using domain ontologies to represent knowledge has shown to be a useful mechanism and the format for managing and exchanging information. Ontologies are the backbone of knowledge representation for the semantic Web. Now, there are so many ontologies on the Internet. Due to the cost and the difficulty of building ontologies, a number of ontology libraries and search engines are emerging to facilitate reusing such knowledge structures. For reducing the cost of building ontology, people always search some candidates from the Internet firstly, and then integrate and refine these candidates to establish their own ontology. But, there are so many candidates that could be found, so the ontology ranking technique is becoming a hot research topic. This paper categorizes the existing ontology ranking methods into two types, and analyzes their characteristics. Then a new ontology ranking method is introduced, which combines the merits of existing algorithms together to improve the experience of ontology searching. It improves the AKTiveRank by selecting a new factor CIM to replace the factor CEM of AKTiveRank. The CIM is a measure to evaluate the importance of one concept in ontology. Finally, two experiments are designed to prove the efficiency of MIDSRank algorithm. The results show that the new method is more efficient.