PatentRank: An Ontology-Based Approach to Patent Search

There has been much research proposed to use ontology for improving the effectiveness of search. However, there are few studies focusing on the patent area. Since patents are domain-specific, traditional search methods may not achieve a high performance without knowledge bases. To address this issue, we propose Patent Rank, an ontology-based method for patent search. We utilize International Patent Classification (IPC) as an ontology to enable computer to better understand the domain-specific knowledge. In this way, the proposed method is able to well disambiguate user's search intents. And also this method discovers the relationship between patents and employs it to improve the ranking algorithm. The empirical experiments have been conducted to demonstrate the effectiveness of our method.