Developing a computational framework for explanation generation in knowledge-based systems and its application in automated feature recognition

TL;DRAbstract

A Knowledge-Based System (KBS) is essentially an intelligent computer system which explicitly or tacitly possesses a knowledge repository that helps the system solve problems. Researches focusing on building KBSs for industrial applications to improve design quality and shorten research cycle are increasingly attracting interests. For the early models, explanability is considered as one of the major benefits of using KBSs since that most of them are generally rule-based systems and the explanation can be generated based on the rule traces of the reasoning behaviors.<br><br>With the development of KBS, the definition of knowledge base is becoming much more general than just using rules, and the techniques used to solve problems in KBS are far more than just rule-based reasoning. Many Artificial Intelligence (AI) techniques are introduced, such as neural network, genetic algorithm, etc. The effectiveness and efficiency of KBS are thus improved. However, as a trade-off, the explanability

Chat with Paper

AI Agents for this Paper