Integrated data-driven DSS in a laboratory environment

Abstract

Decision support technologies have remained individualistic as primarily stand alone platforms. The ability to access and integrate a wide range of such technologies in an Integrated Decision Technology Environment (IDTE) can potentially increase a user's ability to create more complex decision support projects. A well designed IDTE will allow users to identify, learn about, access, execute and integrate disparate decision technologies. Data-Driven DSS provide decision-makers with the capability to store and sort vast amounts of data by leveraging data warehousing and data-mining. These dataoriented decision technologies can assist decision-makers in making better and more informed decisions in shorter durations of time. This thesis focuses on Data-Driven data mining decision technologies and how they can be integrated into an IDTE. In the process of identifying data mining technology requirements, we first created a simple taxonomy characterized by the four categories of association, classification, clustering, and prediction. We then designed a database schema for storing the requisite data about data mining technologies, and case studies illustrating their use. Finally we designed a simple, yet effective, interface for navigating through the data-driven decision technology universe both at NPS and beyond. SQL commands for populating the various screens of the IDTE interface were provided to show proof of concept.