next up previous contents
Next: Multimedia Authoring Environment Up: Expert Systems Previous: Expert Systems

Interactive Learning Environments for Fault Diagnosis

The major bottleneck in the development of such environments has been the acquisition and application of diagnostic knowledge of a chosen device under various faulty conditions. The traditional approach of interviewing human experts for encoding this knowledge is time consuming, may lack co-operation, and the knowledge acquired cannot be applied to other devices and domains. Our research focusses on the automated generation of fault diagnostic knowledge from a description of the structure of a device. The major contributions consist of an integrated knowledge representation methodology for devices, a language for describing device causality, and a meta-knowledge base methodology for automated generation of fault diagnostic knowledge. Most of the terminology and theories used in our methodology is domain-independent, making it applicable for a wide variety of domains and devices with minor modifications.

K. Sandrasegaran, A.S. Malowany

Thierry Baron
Mon Nov 13 10:43:02 EST 1995