Skip to main content

Table 2 Classification accuracy of the proposed algorithm and alternatives on two subsets of the data in the leave-one-out test.

From: Using the information embedded in the testing sample to break the limits caused by the small sample size in microarray-based classification

  BRCA1 BRCA2 PROS PROS-OUT
  DP Others DP Others DP Others DP Others
More correlated 18/18 16.67/18 17/17 16.5/17 59/60 53.3/60 12/15 11.83/15
Less correlated 3/4 1.67/4 4/5 1.67/5 34/41 21.67/41 3/6 0.9/6
  DLBCL-FL ALL-AML I2000   
  DP Others DP Others DP Others   
More correlated 62/62 58.33/62 38/38 34.67/38 58/58 57.83/58   
Less correlated 12/15 9.17/15 0/0 0/0 3/4 1.5/4   
  1. The first subset includes those test feature vectors that are more correlated to the samples of the correct class (called, more correlated in this table). The second subset consists of those test feature vectors that are more correlated to the samples of the incorrect class (referred to as less correlated). The proposed approach is superior in both subsets, but especially so in the less correlated category. This is achieved by taking advantage of the information encoded in the test sample.