Track:
Contents
Downloads:
Abstract:
A Case Based Reasoning (CBR) dominant diagnostic system is presented which collaborates with other intelligent problem solving methodologies. Two of the contributing elements, namely the Matched Vector Functions (MVF) and the Evidence Ratio Factors (ERF) propose an initial decision following their individual data pre-processing. Further, using the Singular Value Decomposition (SVD) method, the proposed solutions are improved and cast as a vote among three main collaborating peers that includes the CBR unit. The paper gives a brief overview of CBR integration in a multi strategy design, and goes further to suggest how to get the most out of intelligent systems that have been so deployed. The main motivation and goal here is performance improvements in the diagnostic accuracy of the system when compared with actual clinician’s diagnostic results, as well as ease of use, and the differential diagnosis feature. Details are discussed in the text and appendix with some results of the capabilities that have benefited our research effort.