Prediction of Success or Failure for Examination using Nearest Neighbor Method to the Trend of Weekly Online Testing


Hideo Hirose


International Journal of Learning Technologies and Learning Environments (IJLTLE), Vol 2, No 1, pp.19-34, May 31, 2019.

Using trends of estimated abilities in terms of item response theory for online testing, we can predict success/failure for term-end examinations for each student at early stages in courses. We applied the newly developed nearest neighbor method for determining the similarity of learning skills in the trends of estimated abilities, resulting in better prediction accuracy for success or failure. This paper shows that the use of the learning analytics incorporating trends for abilities is effective. ROC curve and recall precision curve are also utilized in the proposed method.

Key Words
success/failure prediction, item response theory, nearest neighbor, similarity, online testing, learning analytics.



Times Cited in Web of Science:

Cited in Books: