Analyzing Completeness and Correctness of Utterances Using an ATMS

Authors:
Maxim Makatchev;Kurt VanLehn
Affiliations:
Learning Research and Development Center, University of Pittsburgh;Learning Research and Development Center, University of Pittsburgh
Venue:
Proceedings of the 2005 conference on Artificial Intelligence in Education: Supporting Learning through Intelligent and Socially Informed Technology
Year:
2005

Citing 7
Cited 2

Building problem solvers

Building problem solvers
The Architecture of Why2-Atlas: A Coach for Qualitative Physics Essay Writing

ITS '02 Proceedings of the 6th International Conference on Intelligent Tutoring Systems
Maintaining consistency and plausibility in integrated natural language understanding

COLING '88 Proceedings of the 12th conference on Computational linguistics - Volume 2
Default reasoning in natural language processing

COLING '88 Proceedings of the 12th conference on Computational linguistics - Volume 2
Abductive Theorem Proving for Analyzing Student Explanations to Guide Feedback in Intelligent Tutoring Systems

Journal of Automated Reasoning
The Andes Physics Tutoring System: Lessons Learned

International Journal of Artificial Intelligence in Education
User modeling and problem-space representation in the tutor runtime engine

UM'03 Proceedings of the 9th international conference on User modeling

Combining Bayesian Networks and Formal Reasoning for Semantic Classification of Student Utterances

Proceedings of the 2007 conference on Artificial Intelligence in Education: Building Technology Rich Learning Contexts That Work
Understanding complex natural language explanations in tutorial applications

ScaNaLU '06 Proceedings of the Third Workshop on Scalable Natural Language Understanding

Quantified Score

Hi-index	0.00

Visualization

Abstract

Analyzing coverage of a student's utterance or essay (completeness) and diagnosing errors (correctness) can be treated as a diagnosis problem and solved using a well-known technique for model-based diagnosis: an assumption-based truth maintenance system (ATMS). The function-free first-order predicate logic (FOPL) representation of the essay is matched with nodes of the ATMS that are then analyzed for being within the sound part of the closure or relying on a particular mis-conception. If the matched nodes are sound they are analyzed for representing a particular required physics statement. If they do not represent the required statement, a neighborhood (antecedent and consequent nodes within N inference steps) of these nodes can be analyzed for matching the statement, to give a measure of how far the student utterance is, in terms of a number of inferences, from the desired one.