The Way of Rules Representation in Composited Knowledge Bases

The aim of the paper is to find the proper rules representation in composited knowledge bases (CKB). Such representation lets for improving the efficiency of inference processes. To achieve it we can build a modular, hierarchically organized rule base using cluster analysis method is the way of achieving efficient method of rules representation. There are lot of clusters representations, but the most often used is the method based on centroid. If descriptive methods are not sufficient then it is possible to use methods based on visualisation of rules - PCA. With minimal additional effort it provides a roadmap for how to reduce a complex rule set to a lower dimension to reveal the sometimes hidden, simplified structure that often underlies it.