A Hypergraph-based Method for Pharmaceutical Data Similarity Retrieval

Drug and compound similarity retrieval is very important for new drug research and development. Most pharmaceutical database provide keyword searching service. Because keyword search method cannot identify entity semantic similarity information, so the retrieval results often got poor drug or compound semantic similarity. In this paper, we propose an attribute semantic similarity retrieval method for pharmaceutical data based on hypergraph and natural language processing technology. Firstly, we use natural language processing technology to research and construct the drug attribute semantic similarity network. Then, we continue building a hypergraph based on drug attribute to get better retrieval efficiency. The experimental results show that, our method can provide similarity retrieval service for researchers.