You are here:

Automatic Annotation Method on Learners' Opinions in Case Method Discussion
ARTICLE

, ,

Interactive Technology and Smart Education Volume 12, Number 2, ISSN 1741-5659

Abstract

Purpose: The purpose of this paper is to annotate an attribute of a problem, a solution or no annotation on learners' opinions automatically for supporting the learners' discussion without a facilitator. The case method aims at discussing problems and solutions in a target case. However, the learners miss discussing some of problems and solutions. Design/Methodology/Approach: Because opinions about problems and solutions on the same case are similar to each other, the proposed method uses opinions that are correctly annotated in past discussions for annotating an appropriate attribute on each opinion in discussions of the same case. The annotation on each opinion is identified by Support Vector Machine learned with opinions and annotations in the past discussion. Findings: Compared to a simple method that uses decision tree classification, this proposed method improves the recall rate and the precision rate of annotating the attribute by over 10 per cent. The proposed method is effective for automatic annotation. Originality/Value: Because the recall rate and the precision rate of annotating an attribute of a problem are over 80 per cent, it is possible to make learners aware of problems that they should discuss. On the other hand, the recall rate and the precision rate of annotating an attribute of a solution are still low. The authors discuss the research issue to improve the rates for automatic annotation.

Citation

Samejima, M., Hisakane, D. & Komoda, N. (2015). Automatic Annotation Method on Learners' Opinions in Case Method Discussion. Interactive Technology and Smart Education, 12(2), 90-99. Retrieved July 14, 2020 from .

This record was imported from ERIC on November 3, 2015. [Original Record]

ERIC is sponsored by the Institute of Education Sciences (IES) of the U.S. Department of Education.

Copyright for this record is held by the content creator. For more details see ERIC's copyright policy.

Keywords