Application of artificial neural networks for prediction of learning performances
© 2016 IEEE. Artificial neural networks (ANNs) have rarely been used in the field of medical education, especially in the prediction of learning performances. This study aims to evaluate the potential application of ANN models for predicting learning performances, in comparison with multivariate log...
Saved in:
| Main Authors: | , |
|---|---|
| Other Authors: | |
| Format: | Conference or Workshop Item |
| Published: |
2018
|
| Subjects: | |
| Online Access: | https://repository.li.mahidol.ac.th/handle/123456789/43424 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Mahidol University |
| Summary: | © 2016 IEEE. Artificial neural networks (ANNs) have rarely been used in the field of medical education, especially in the prediction of learning performances. This study aims to evaluate the potential application of ANN models for predicting learning performances, in comparison with multivariate logistic regression models. The predictor variables included demographics, high-school backgrounds, first-year grade-point averages, and composite scores of examinations during the course. Medical student learning performances were represented by their normalized T-scores of the total examination score. Three ANN models, including a support vector machine, were used to predict performance. A comparison between the models, based upon areas under the receiver operating characteristic curve values, showed no significant differences between the ANNs and logistic regression models (p > 0.05 for all pairs in the comparison). This work thus reveals the promising potential for the application of ANNs in the prediction of learning performances, in the field of medical education. |
|---|