Figure 1
From: Multi-view singular value decomposition for disease subtyping and genetic associations
Comparison of different methods on AUC values in the simulation. The box plot of AUC values obtained from all approaches in the comparison is shown for the simulated data. The methods were: A1 - the proposed method, A2 - single-view SSVD, A3 - co-regularized spectral clustering, A4 - kernel addition, A5 - kernel product, and A6 - feature concatenation. The parameter e reflects the level of genotypic effect to the phenotypic variation in the simulated data. The AUC values characterize the genetic separability of the clusters resulting from each method.