Study on Adverse Events ofDrugs usingData Mining

使用FDA不良事件报告系统(AERS)构建的数据库，利用数据挖掘的方法，找到一些更容易导致Stevens-Johnson综合征(SJS)和中毒性表皮坏死松解症(TEN)的药物。特别是，我们将关联规则应用于每个患者的药物组合。作为挖掘的结果，我们构建了一个新的过程来寻找重要的可疑药物。此外，我们还揭示了可疑药物中经常存在的一些化学特征。

The database constructed by the FDA Adverse Event Reporting System (AERS) was used in order to find some drugs which are more likely to causeStevens - Johnson syndrome (SJS) and toxic epidermal necrolysis (TEN) using data mining methods. Especially, we utilized association rule applying to the combinations of drugs in each patient. As a result of the mining, we constructed a new process to find significant suspect drugs. In addition we also revealed some chemical features which frequently exist in the suspect drugs.