Knowledge base design enhancement through reliability estimation

摘要

Many knowledge-based systems are designed and built with little attention paid to their reliability. Failure to perform reliability measurement and analysis means that fielded systems might fail more frequently than expected, customers might not be willing to trust knowledge-based systems for critical tasks, and developers may never fully understand the capabilities of the knowledge-based technology they are using. Knowledge-based systems are not subject to reliability measurement and analysis primarily because the rapid prototyping style of development makes it difficult to formally test and model the results of testing to produce an accurate estimate of the reliability. This paper presents an approach to reliability prediction of a knowledge-based system during any stage of the rapid prototyping development process. We illustrate the approach using a system developed to provide guidance on materials selection for varying climatic conditions.