Research Article
Evolutionary Feature Selection for Big Data Classification: A MapReduce Approach
Table 4
AUC results for the Spark classifiers using epsilon.
| Threshold | Features | Logistic Regression | Naive Bayes | SVM () | SVM () | Training | Test | Training | Test | Training | Test | Training | Test |
| 0.00 | 2000 | 0.6786 | 0.6784 | 0.7038 | 0.7008 | 0.6440 | 0.6433 | 0.6440 | 0.6433 | 0.55 | 721 | 0.6985 | 0.7000 | 0.7154 | 0.7127 | 0.6855 | 0.6865 | 0.6855 | 0.6865 | 0.60 | 337 | 0.6873 | 0.6867 | 0.7054 | 0.7030 | 0.6805 | 0.6799 | 0.6805 | 0.6799 | 0.65 | 110 | 0.6496 | 0.6497 | 0.6803 | 0.6794 | 0.6492 | 0.6493 | 0.6492 | 0.6493 |
|
|