Conception of an intelligent tutoring system in cost engineering: Knowledge representation, pedagogical interactions, and system operation.

摘要

Cost engineering is a domain taught to university students in all engineering disciplines. Its goal is to provide engineers with mathematical analysis tools useful to make economically accurate decisions. This discipline is complex and raises several learning difficulties among students, which brings the need for, in addition to human assistance, a computer-based instruction system. However, to efficiently fix difficulties and misunderstandings encountered by students, such a tutoring system must have strong pedagogical skills in addition to domain-related knowledge. Moreover, cost engineering is a domain that involves not only theoretical knowledge, but also problem-solving skills. Finally, there exist no tutoring systems yet in cost engineering.;This thesis presents an intelligent tutoring system (ITS) in cost engineering. We analyse in detail the representation of all involved types of knowledge: domain knowledge, problem-solving knowledge, and tutoring knowledge. We also exhibit some interesting issues of our knowledge representation, in particular four generic operating modes of an ITS applied to a domain involving problem-solving skills. Finally, we show our design for implementing that system.