Selecting critical features for data classification based on machine learning methods

Journal of Big Data

Table 11 Statistics by class of Different Classifiers with Car Evaluation Dataset (4 features)

Evaluation	Class: acc	Class: good	Class: unacc	Class: vgood	Evaluation	Class: acc	Class: good	Class: unacc	Class: vgood
RF + SVM Accuracy: 0.8401					RF + LDA Accuracy: 0.843
Precision	0.6095	NA	0.9646	0.53846	Precision	0.6111	NA	0.9731	0.53846
Recall	0.8421	0.00000	0.9008	0.53846	Recall	0.8684	0.00000	0.8967	0.53846
RF + KNN Accuracy: 0.8227					RF + RF Accuracy: 0.9336
Precision	0.5810	0.000000	0.9518	0.55556	Precision	0.9054	0.55000	0.9832	0.75000
Recall	0.8026	0.000000	0.8967	0.38462	Recall	0.8816	0.84615	0.9669	0.69231