Making decisions in a two-stage identification system with knowledge updating

In this article, an algorithm for updating knowledge in an adaptive system to support decision making is proposed. Taking into account the specific character of the process, a two-stage identification approach is used. The first stage is built for diagnostic purposes, i.e., the estimated parameters in the first-stage relationship are used to make a decision at the second stage. The proposed algorithms to support decision making at the second stage rest on knowledge extracted from human experts and the effects of the diagnosis (based on identification results) at the first stage. To examine the effectiveness of the proposed algorithm, a numerical simulation of a biomedical problem is carried out.