Table 3 Average classification accuracy and associated standard deviations(in parentheses) of prostate cancer subtypes in the test group using SBGG, BBGDE, SVM and Random Forest models for using 10 marker genes
From: A Bayesian approach for inducing sparsity in generalized linear models with multi-category response
Sample Type | SBGG | SBGDE | SVM | Random Forest |
|---|---|---|---|---|
Benign | 89.4 (6.1) | 95.1 (6) | 84.4 (5.3) | 91.1 (4.5) |
PIN | 62.5 (1.6) | 61.7 (2.8) | 9 (7.2) | 61.4 (1.9) |
PCA | 98.7 (0.7) | 86.9 (1.1) | 37.4 (9) | 86.7 (2.1) |
MET | 59.4 (2.06) | 56 (3.2) | 55.3 (1.2) | 82.8 (7.3) |