Early detection of diseases using electronic health records data and covariance-regularized linear discriminant analysis. In Proceedings of the IEEE Embs International Conference on Biomedical and Health Informatics, Orlando, FL, USA, 16-19 February 2017....
In addition to AUC, we also evaluated the learning curve of each model. While the AUC of Random Forest was slightly lower than Stacking, its convergence in the shape of learning curves (training and testing) was superior to both Stacking and Voting. Following multidisciplinary team discussion, ...