Skip to main content

Table 1

From: Combining dissimilarity based classifiers for cancer prediction using gene expression profiles

 

% Error

%False Negative

Method

Breast

Leukemia

Breast

Leukemia

Euclidean

10.2%

6.9%

4%

6.94%

Cosine

14.2%

1.38%

4%

1.38%

Correlation

14.2%

2.7%

6.1%

2.7%

2

12.2%

1.38%

4%

1.38%

Manhattan

12.2%

5.5%

4%

4.16%

Spearman

16.3%

8.3%

6.1%

5.5%

Kendall-Tau

18.3%

8.3%

6.1%

5.5%

Bagging

6.1%

2.77%

2%

1.38%

Combination

8.1%

1.38%

2%

1.38%

  1. Experimental results for the ensemble of SVM classifiers. Classifiers based solely on a single dissimilarity and Bagging have been taken as reference.