Supporting Discovery in Medicine by Association Rule Mining of Bibliographic Databases

The paper presents an interactive discovery support system for the field of medicine. The intended users of the system are medical researchers. The goal of the system is: for a given starting concept of interest, discover new, potentially meaningful relations with other concepts that have not been published in the medical literature before. We performed two types of preliminary evaluation of the system: 1) by a medical doctor and 2) by automatic means. The preliminary evaluation showed that our approach for supporting discovery in medicine is promising, but also that some further work is needed, especially on limiting the number of potential discoveries the system generates.

• Type: Conference
• Year: 2000
• Where: PKDD
• Authors: Dimitar Hristovski, Saso Dzeroski, Borut Peterlin, Anamarija Rozic-Hristovski
• Keywords: Data Mining, Interactive Discovery Support, Medical Researchers, PKDD 2000, Preliminary Evaluation