Database application model and its service for drug discovery in Model-driven architecture

Abstract

Big data application has many data resources and data. In the first stage of software engineering, a service overview or a system overview cannot be seen. In this paper, we propose that two processes of “Big data analytics” and “Implementation of data modeling” should be collaborated with Model-driven architecture (MDA). Data modeling with those two process in MDA should be repeated fast in order to verify the data model and to find a new data resource for a service. Our first research goal of big data application is to predict side effect of drug which is one of screening methods in drug discovery. This prediction model is constructed with data mining methods at the intersection of statistics, machine learning and database system. Moreover, a new service for drug discovery by new uses for old drugs can be found in data modeling and developed. We demonstrate that the prediction model and the data model for drug discovery are implemented as a prototype system to verify those models and their practicality.