Enhanced Active Learning in Developing Highly Interpretable Decision Support System

Developing highly interpretable decision rules commonly presents significant challenges to decision support system. In previous research work, partial information had provided complex decision in the problem of learning classifiers. The behaviour of some learning algorithm may only be explored by uncertainty analyses. We propose a novel information extraction by utilizing weighted active learning fuzzy measure based on objective function to quantify the goodness of cluster models that comprise prototypes and data partition in decision modelling. By choosing appropriate weights for pre labelled data, the nearest neighbour classifier consistently improves on the original classifier.Keywords Uncertainty, Decision Support System, Fuzzy Cluster Analysis