Publications (3)0 Total impact
-
[show abstract]
[hide abstract]
ABSTRACT: The major focus of recent knowledge acquisition research has been on problem-solving methods (PSM). This paper present results where a PSM developed for classification has been extended to handle a configuration or parametric design task, designing ion chromatography methods in analytical chemistry. Surprisingly good results have been obtained seemingly because any knowledge that has been added to the knowledge base, has been added precisely to overcome any limitations of the PSM. These results suggest a trade-off between domain knowledge and the power of the PSM and that greater use of domain knowledge would facilitate re-use by allowing PSMs to be used for a broader range of tasks. INTRODUCTION The critical insight with the emergence of knowledge based systems (KBS), was the usefulness of knowledge as compared to pure search. This was formulated in ideas such as the "knowledge principle". This resulted in expert system shells which while specifically intended for the accumulation o...
10/2000;
-
[show abstract]
[hide abstract]
ABSTRACT: Ripple Down Rules (RDR) is a knowledge acquisition method for knowledge based systems (KBS) which facilitates incremental acquisition of knowledge and ensures that the previous performance of the KBS is not degraded by the incremental addition of the new knowledge. This approach is now well established for single classification tasks and more recently has been extended to multiple classification tasks. This paper describes the further extension of the approach to configuration tasks. The test domain for this study is the configuration of ion chromatography methods in analytical chemistry. 1. RDR Background Ripple Down Rules (RDR) is based on the idea that when a KBS makes an incorrect conclusion the new rule that is added to correct that conclusion should only be used in the same context in which the mistake was made(Compton and Jansen 1990). In practice this means attaching the rule at the end of the sequence of rules that were evaluated leading to the wrong conclusion. Thus, this r...
07/1999;
-
[show abstract]
[hide abstract]
ABSTRACT: A classifier system based on genetic algorithm methodology was developed for the automatic extraction of production rules from a database of about 6000 ion chromatography (IC) method examples. This machine learning strategy generated heuristics that can assist in the choice for a detection method for a specified set of IC method and solute properties. It was shown that the final set of rules proposed detectors that agreed with the database for 76% of the cases. Application to a separate test set showed a prediction ability of 82%. The database, because of the characteristics of the included cases, did not allow for a significant improvement of these results. However, the results are of significance for the further development of knowledge systems, which assist in the design of IC methods. Furthermore, this dataset comprised a considerable challenge to the applied machine learning method.
Analytica Chimica Acta.
