Predicting drug side effects based on link prediction in bipartite network

To pursue a new approach for discovering new side effects of drugs, we construct a drug-side effect network based on the SIDER2 database, and introducing link prediction methods in the network. We develop and evaluate the framework for drug side effect prediction based on link prediction in bipartite network. Integrating side effect appearance frequency has also been proved to be of value in improving the performance of drug side effect prediction.