Building knowledge bases through multistrategy learning and knowledge acquisition

This paper presents a new approach to the process of building a knowledge-based system which relies on a tutoring paradigm rather than traditional knowledge engineering. In this approach, an expert teaches the knowledge based system in much the same way the expert would teach a human student, by providing specific examples of problems and solutions, explanations of these solutions, or supervise the system as it solves new problems. During such interactions, the system extends and corrects its knowledge base until the expert is satisfied with its performance. Three main features characterize this approach. First, it is based on a multistrategy learning method that dynamically integrates the elementary inferences that are employed by the single-strategy learning methods. Second, much of the knowledge needed by the system is generated by the system itself. Therefore, most of the time, the expert will need only to confirm or reject system-generated hypotheses. Third, the knowledge base development process is efficient due to the ability of the multistrategy learner to reuse its reasoning process, as well as the employment of plausible version spaces for controlling the knowledge base development process. This paper illustrates a cooperation between a learning system and a human expert in which the learner performs most of the tasks and the expert helps it in solving the problems that are intrinsically difficult for a learner and relatively easy for an expert.