Journal of ARTIFICIAL INTELLIGENCE IN EDUCATION
Volume 7, Number 1 1996
Articles
Individual Selection of Examples in an Intelligent Learning Environment
Gerhard WeberLaw Encoding Diagrams for Instructional Systems
Peter C.-H. ChengRefinement-Based Student Modeling and Automated Bug Library Construction
Paul Baffes and Raymond Mooney
Abstracts
Individual Selection of Examples in an Intelligent Learning Environment
GERHARD WEBER
FB I-Psychology
University of Trier
D-54286 Trier, Germany
Learning from examples is one of the main sources of acquiring procedural skills. Carefully chosen examples have a high impact on learning to solve problems (Reed & Bolstad, 1991). In large and complex problem solving domains it will not be possible to provide an optimal example in the textbook for all tasks learners have to solve during a course. Therefore, it will be challenging for complex learning environments to have a mechanism that dynamically selects the most useful examples from an individual learning history. In the intelligent programming environment ELM-PE that supports beginners in learning the programming language LISP, an explanation-based retrieval method (EBR) is used to retrieve and provide examples and individual remindings automatically. The EBR method retrieves organizationally similar cases from an individual episodic case base. In empirical studies, it has been shown that the EBR method is able to offer analogies that are better suited than preselected examples from the textbook and in many cases they are better suited than examples that students selected by themselves.
Law Encoding Diagrams for Instructional Systems
PETER C.-H. CHENG
ESRC Centre for Research in Development, Instruction and Training
University of Nottingham, University Park
Nottingham, NG7 2RD, UK
This paper introduces the concept of Law Encoding Diagrams, LEDs, a novel class of diagrammatic knowledge representations. A LED is a representation that correctly encodes the underlying relations of a law, or a system of simultaneous laws, in the structure of a diagram by the means of geometric, topological and spatial constraints, such that each instantiation of a single diagram represents an instance of the phenomenon or one case of the laws. Different forms of inference are supported by LEDs, ranging from simple reasoning to complex conceptual problem solving. A discovery learning environment with LEDs for perfectly elastic collisions is described. A small scale empirical study of problem solving and learning on the system has been conducted with university students. They were given brief training on the system and no instruction in problem solving strategies with LEDs, but they were able to use the system to solve a variety of problems. In post-tests half the subjects used LEDs for problem solving with novel problem solving strategies, in contrast to their own pre-test approaches and the approaches of the other subjects. The LED users have a better understanding of the diagrammatic constraints of the LEDs, which apparently comes from their more comprehensive examination of the possible structural forms of the LEDs.
Refinement-Based Student Modeling and Automated Bug Library Construction
PAUL BAFFES
SciComp, Inc., 5806 Mesa Drive
Suite 250, Austin, TX 78731, USA
RAYMOND MOONEY
Department of Computer Sciences, Taylor Hall 2.124,
The University of Texas at Austin, Austin, TX 78712, USA
A critical component of model-based intelligent tutoring systems is a mechanism for capturing the conceptual state of the student, which enables the system to tailor its feedback to suit individual strengths and weaknesses. To be useful such a modeling technique must be practical, in the sense that models are easy to construct, and effective, in the sense that using the model actually impacts student learning. This research presents a new student modeling technique which can automatically capture novel student errors using only correct domain knowledge, and can automatically compile trends across multiple student models. This approach has been implemented as a computer program, ASSERT, using a machine learning technique called theory refinement, which is a method for automatically revising a knowledge base to be consistent with a set of examples. Using a knowledge base that correctly defines a domain and examples of a student's behavior in that domain, ASSERT models student errors by collecting any refinements to the correct knowledge base which are necessary to account for the student's behavior. The efficacy of the approach has been demonstrated by evaluating ASSERT using 100 students tested on a classification task covering concepts from an introductory course on the C++ programming language. Students who received feedback based on the models automatically generated by ASSERT performed significantly better on a post test than students who received simple reteaching.