Towards on-line ADR detection: an innovative data cube-based approach.

The problem of Adverse Drug Reactions (ADRs) detection is a key issue in the pharmacovigilance community. Although many related researches have been proposed in recent years, majority of them were concerned about the accuracy of the detected adverse drug reactions, very few paying attention to the performance issue of how to quickly accomplish the detection process. In this work, we propose the concept of multivalued-dimension contingency cube (MDC-cube) to facilitate multidimensional, fast and online detection of ADRs, especially those caused by drug interactions. Experiments conducted on the FDA AERS data set show that our MDC-cube based method is an order of magnitude faster than the state-of-the-art ABCM-MS method.