| The main purpose of this research is to use selected tools and methodologies from the distributed artificial intelligence (DAI) field to create a functioning knowledge-based decision support system (KBDSS). This KBDSS performs telephone analysis and financial analysis. Furthermore, an observational study is conducted to compare the prototype of the KBDSS, which we call Sherpa, to the system currently in use.; The architecture of the prototype comprises a set of problem solvers, all operating under the guidance of meta-level knowledge. Each problem solver works as a classifier system. The problem solving knowledge of each solver is developed using machine learning and/or knowledge engineering.; As part of the observational study, variables are identified and hypotheses formulated. Statistical analysis is carried out to test the hypotheses. Sherpa outperforms the currently used system in all aspects tested. |
