LCT の結果を使った期末試験の合否予測:2017LCT から作成したモデルに2018LCT を適用

URI http://harp.lib.hiroshima-u.ac.jp/it-hiroshima/metadata/12342
ファイル
タイトル
LCT の結果を使った期末試験の合否予測:2017LCT から作成したモデルに2018LCT を適用
別タイトル
Success/Failure Prediction for Final Examination Using the LCT Results:A Model using 2017 LCT Data is Applied to 2018 LCT Data
著者
氏名 廣瀬 英雄
ヨミ ヒロセ ヒデオ
別名 HIROSE Hideo
キーワード
online testing
learning check testing
item response theory
drop-out
抄録

The author proposed a prediction method to find students with drop-out risks as early as possible
by using the learning check testing (LCT) results, where the nearest-neighbor method was used in
the previous paper. If we want to build a mathematical model using a part of 2018 LCT results and
we want to predict 2018 end-term results using the model, it is impossible to do that in the middle
of the term, because we will not have the final results until the end of the term. Thus, in this paper,
we investigated whether the proposed method was valid or not by applying the model built using
2017 LCT data to newly obtained 2018 LCT results. Using 2017 model, the prediction accuracies and
the corresponding confusion matrices applying to 2018 case were almost the same as that applying
to 2017 case. Therefore, the previously proposed method is proved to be useful.

掲載雑誌名
広島工業大学紀要. 研究編
54
開始ページ
137
終了ページ
140
出版年月日
2020-02
出版者
広島工業大学
ISSN
1346-9975
NCID
AA11599110
本文言語
日本語
資料タイプ
紀要論文
著者版フラグ
出版社版
権利情報
publisher
区分
it-hiroshima