Academic
Publications
Diagnosing Natural Language Answers to Support Adaptive Tutoring

Diagnosing Natural Language Answers to Support Adaptive Tutoring,Myroslava Dzikovska,Gwendolyn E. Campbell,Charles B. Callaway,Natalie B. Steinhauser,

Diagnosing Natural Language Answers to Support Adaptive Tutoring   (Citations: 6)
BibTex | RIS | RefWorks Download
Understanding answers to open-ended explanation questions is important in intelligent tutoring systems. Existing systems use natural language techniques in essay analysis, but revert to scripted interaction with short-answer questions during remediation, making adapting dialogue to individual students difficult. We describe a corpus study that shows that there is a re- lationship between the types of faulty answers and the remediation strategies that tutors use; that human tu- tors respond differently to different kinds of correct an- swers; and that re-stating correct answers is associated with improved learning. We describe a design for a di- agnoser based on this study that supports remediation in open-ended questions and provides an analysis of natu- ral language answers that enables adaptive generation of tutorial feedback for both correct and faulty answers.
Cumulative Annual
View Publication
The following links allow you to view full publications. These links are maintained by other sources not affiliated with Microsoft Academic Search.
    • ...The output from the parser is passed on to the domain reasoning and diagnosis components [8,9] to check the validity of the student’s explanation...

    Myroslava O. Dzikovskaet al. Intelligent Tutoring with Natural Language Support in the Beetle II Sy...

    • ...Student input may be restricted to short answers to questions (single words or phrases) [4,5] 1 , support a small set of longer sentences [6,7,8,9,10], or attempt to interpret extended answers to “Why” questions [8,11,12]...
    • ...The feedback may be completely hand-authored [4,5,6,8], or generated automatically based on the system’s internal representation [7,9,10,12]...
    • ...When the student answered a question correctly, the ITS confirmed the correct answer, occasionally restating it with improved terminology [12]...

    Myroslava O. Dzikovskaet al. Content, Social, and Metacognitive Statements: An Empirical Study Comp...

    • ...In our own prior work, for example, we analyzed dialogue behaviors and learning using corpora of computer and human tutoring spoken dialogues in the conceptual mechanics domain [4,5], as well as human tutoring typed dialogues in the electricity domain [6]...
    • ...The corpus was collected during a study comparing the use of different forms of tutor questions (openended versus short answer versus multiple choice) [6]...
    • ...We are currently examining whether our prior results regarding tutor behaviors (e.g., questioning [5] and restating [6]) also generalize across our corpora, and plan to extend our automatic content tagging to use more sophisticated natural processing techniques...

    Diane J. Litmanet al. Using Natural Language Processing to Analyze Tutorial Dialogue Corpora...

    • ...1 Answers to explanation questions are hand-coded by tutors because they are not always required to be logically complete (Dzikovska et al., 2008)...
    • ...And then?” We can do this because we use a diagnoser that explicitly identifies the correct objects and relationships in the answer (Dzikovska et al., 2008), and we are using a deep generation system that can take those relationships and automatically generate a rephrasing of the correct portion of the content...

    Myroslava Dzikovskaet al. Dealing with Interpretation Errors in Tutorial Dialogue

Sort by: