An improved clear cell renal cell carcinoma stage prediction model based on gene sets

BMC Bioinformatics

Table 2 The performance of machine learning based-models developed using FLJ-set of 23 selected features on the training set with 10-fold cross-validation set and independent testing set for gene data without discretization

Algorithms	Methods	Performance Measures on test set
Algorithms	Methods	Sensitivity	Specificity	Accuracy(%)	MCC	AUC
Logistic Regression	10-fold	0.750	0.805	78.45	0.556	0.855
Logistic Regression	Testing	0.756	0.767	77.87	0.554	0.860
SVM	10-fold	0.680	0.868	79.27	0.562	0.852
SVM	Testing	0.714	0.877	81.15	0.603	0.860
MLP	10-fold	0.706	0.828	77.83	0.508	0.840
MLP	Testing	0.776	0.836	81.15	0.609	0.858
Naive Bayes	10-fold	0.695	0.820	77.17	0.519	0.828
Naive Bayes	Testing	0.735	0.836	79.51	0.572	0.819
Random Forest	10-fold	0.499	0.866	71.75	0.398	0.764
Random Forest	Testing	0.612	0.863	76.23	0.496	0.828

ISSN: 1471-2105