A Performance Weighted Collaborative Filtering algorithm for personalized radiology education

- A performance weighted prediction algorithm is proposed to predict the difficulty level of each case for each trainee.
- The algorithm assigns an optimal weight to each rating of each trainee to make prediction.
- The assigned weight to a rating is related to the performance level during which the rating was made.
- Experiment results show that the algorithm outperforms traditional collaborative filtering algorithms in terms of MAE.
- The algorithm is a promising method to development of personalized training system in radiology education.