Figure 3. R-precision obtained in the disease tagging experiment by the different patient representations over several prediction time intervals (expressed as number of days).

We reports results for patients represented with original descriptors (RawFeat) and pre-processed by principal component analysis (PCA), independent component analysis (ICA), Gaussian mixture model (GMM), k-means clustering (K-Means), and three-layer stacked denoising autoencoders (DeepPatient).