Skip to main content
. Author manuscript; available in PMC: 2019 Jun 13.
Published in final edited form as: Proc Int World Wide Web Conf. 2019 May;2019:659–669. doi: 10.1145/3308558.3313548

Table 3:

Table of AUC, running time, and average overlapping using different methods. The highest AUC value means extracted phenotypes have stronger discrimination. The lowest running time indicates our distributed method can significantly accelerate the computation time. Compared to CP-APR and FlexiFact, adding angular penalty improved the distinction significantly.

Model AUC Time Avg Overlap
CP-APR [8] 0.63 > 1 hour 0.3
FlexiFact [6] 0.65 35 mins 0.37
Granite [16] 0.67 > 1 hour 0.3
SGranite (β1 = 0) 0.68 20 mins 0.1
SGranite (β1 > 0) 0.71 25 mins 0.07