Skip to main content

Table 2 A comparison of different imputation methods

From: Optimization and expansion of non-negative matrix factorization

 BaselineMediansMICEMissForestNMF
MSE4.42720.52290.99500.41750.4191
MKL0.31660.03890.06880.02980.0301
Time (Sec.)0.00000.000090.267042.40100.1400
  1. Imputations on a subset of NSCLC microarray data, which composes 200 genes and 100 samples. 30% of the entries are randomly deleted, i.e., missed. MSE = mean square error, MKL = Mean KL-divergence distance and Time = user time