Water pressure tests are a common method of soil exploration in engineering geology. The classification of water pressure test results is usually carried out by expert geologists. This paper presents a method for the automatic classification of water pressure test curves. After discussing the features extracted from the test curves the fuzzy rule base used for classification is described. Some ...

This paper presents a uni ed methodology for reasoning in classi cation systems. The methodology is based on a two-stage structure that incorporates both neural and Bayesian formulations in the rst stage and a rule-based system created by extracting rules from both the classi ers in the second stage. The rule-based system provides a measure of the cause-e ect relationship between the inputs and...

The support of manufacturing machines with IT solutions becomes state-of-the art in the production process. For this reason the European Chinese research project KoBaS (Knowledge Based Customized Services for Traditional Manufacturing Sectors Provided by a Network of High Tech SMEs) aims to create customized services for manufacturing machines providing a set of tools including advanced task an...

This paper considers the automatic design of fuzzy rule-basedclassification systems based on labeled data. The classification performance andinterpretability are of major importance in these systems. In this paper, weutilize the distribution of training patterns in decision subspace of each fuzzyrule to improve its initially assigned certainty grade (i.e. rule weight). Ourapproach uses a punish...

We propose a new method to use an incom plete rule base with imprecize descriptions of variables We extend classical interpola tive reasoning to this case under the as sumption of graduality in variations of the variables by using a linguistic approach

We consider the requirements that a citation system must fulfill in order to cite structured and evolving data sets. Such a system must take into account variable granularity, context and the temporal dimension. We look at two examples and discuss the possible forms of citation to these data sets. We also describe a rule-based system that generates citations which fulfill these requirements.