Similarity, Uncertainty and Case-Based Reasoning in Patdex

Patdex1 is an expert system that uses case-based reasoning in the diagnosis of faults of complex machines. It is integrated in the Moltke workbench2 for technical diagnosis, which was developed at the University of Kaiserslautern in recent years (cf., e.g., [4, 5, 23]); Moltke contains other parts as well (cf., e.g., [16]), in particular a model-based approach (cf. [21, 22]); it is essentially in Patdex [3] that the heuristic features are located. The use of cases also plays an important role in knowledge acquisition. In this paper we present the underlying principles of Patdex and embed its main concepts into a theoretical framework. This research has a number of mainly indirect connections to the work of Woody Bledsoe. We mention his interest in analogy, his early connectionist work, and his influence in merging mathematics and artificial intelligence. For the first author the main point to note is that Woody Bledsoe brought him into contact with AI at an early stage. More than twenty years ago we started a lively discussion that still goes on and that will, it is hoped, last for many more years.