Education and demonstration – Question or problem eliciting response
Reexamination Certificate
2001-06-19
2003-10-21
Vo, Hieu T. (Department: 3747)
Education and demonstration
Question or problem eliciting response
C704S270000, C704S255000
Reexamination Certificate
active
06634887
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention generally relates to methods and systems for instructing or tutoring a student. The present invention more particularly relates to methods and systems for student instruction that employ an intelligent tutoring system for initiating a dialog between the tutoring system and the student.
2. Description of the Related Art
Some existing computer-based systems employ a model-tracing approach to adapt instruction to an individual student. Systems that apply model-tracing provide just instructional messages either in response to an incorrect step or when a student is stuck and requests a hint. While error messages adapt to different kinds of errors, the hint messages a student may request in a particular context do not adapt to prior student errors.
Traditional model-tracing tutors provide “buggy feedback” in the form of messages based on a diagnosed error and provide one or more hints that direct a student to a correct action or response. With regard to buggy feedback, each production rule in the student model portion of the tutoring system that represents a common error is marked as a buggy rule. These production rules also contain templates to generate a feedback message. One such buggy rule, for example, might have a template that tells the student to put parentheses around a portion of an incorrect student response for which parentheses are required. Hints are usually only given when requested by the student. When a student asks for a hint on a given question, the tutor executes a production system to compute a next step and fill a template to generate a hint message.
There are several disadvantages to generating feedback using such buggy feedback and hints. Foremost, it is generally preferable to ask the student a question that requires reflection rather than reveal a portion of the solution to the problem. Conventional model-tracing systems may ask questions of the student but typically do not allow a response, than therefore the questions are merely rhetorical in nature. It is believed that student learning can be enhanced by having the tutoring system ask new, related questions, rather than just provide more information. For a student having difficulty solving a problem, the tutoring system should decompose the problem and ask the student questions about the goals that the student did not accomplish correctly as evidenced by incorrect responses. Systems that generate feedback only in this manner lack a feel of engaging in a personal dialog during a tutoring session.
What are needed, therefore, that are not provided by conventional tutoring systems, are intelligent tutoring methods and systems that leverage interactive dialog to facilitate student comprehension in connection with solving problems. Tutorial methods and systems are needed that can simulate the teaching strategies and actions of human tutors. The improved tutoring methods and systems should establish a coherent dialog that includes sub-dialogs that ask new questions without merely providing straightforward hint information. Tutoring methods and systems are needed that can recognize portions of a correct answer, provide positive feedback on those portions, and also tutor the student on the incorrect portions of the answer. The improved tutoring methods and systems should support a plurality of different tutorial strategies.
SUMMARY OF THE INVENTION
Methods and systems are provided for tutoring a student in solving a problem described in the form of a question-and-answer dialog with the student. The method employs a student model for diagnosing student input and a tutorial model for deciding what new questions to plan to ask the student. The method comprises the following steps once student input has been diagnosed by the student model: receiving the diagnosis in a tutorial processing module; generating an agenda including at least one question; applying at least one tutorial strategy to the diagnosis; and, providing feedback to the student based on application of the tutorial strategy and the agenda to the diagnosis. In certain aspects of the method, the tutorial strategy includes at least one knowledge remediation dialog and/or at least one knowledge construction dialog that can be applied to generate dialog between the tutor and the student. System embodiments are also provided that incorporate the functions of the methods for tutoring a student. In addition, in other embodiments, the present tutoring methods and systems are practiced in connection with instructions stored on a computer-readable medium.
REFERENCES:
patent: 5920838 (1999-07-01), Mostow et al.
patent: 2002/0156632 (2002-10-01), Haynes et al.
Presentation by Ken Koedinger: Social Science Data Infrastructure Conference. New York University. “Learning Factors Analysis: Mining Student-Tutor Interactions to Optimize Instruction”; Nov. 12-13, 1999.
Presentation by Ken Koedinger: Center for Interdisciplinary Research on Constructive Learning Environments (CIRCLE) Seminar. LRDC, University of Pittsburgh. “Supporting Meta-Cognitive Knowledge Construction: Steps Toward Third Generation Intelligent Tutors for Mathematics”; Sep. 24, 1999.
Neil T. Heffernan III, “Intelligent Tutoring Systems have Forgotten the Tutor: Adding a Cognitive Model of Human Tutors”. School of Computer Science, Carnegie Mellon University, Pittsburgh, Pennsylvania, Mar. 2001.
Declaration of Kenneth R. Koedinger, Ph.D. executed May 9, 2003, 32 pp. (including exhibits).
Declaration of Neil T. Hefferman, Ph.D., executed May 8, 2003, 32 pp. (including exhibits).
Heffernan, III Neil Thomas
Koedinger Kenneth R.
Carnegie Mellon University
Kirkpatrick & Lockhart LLP
Vo Hieu T.
LandOfFree
Methods and systems for tutoring using a tutorial model with... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Methods and systems for tutoring using a tutorial model with..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Methods and systems for tutoring using a tutorial model with... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3113793