Classification of premalignant pancreatic cancer mass-spectrometry data using decision tree ensembles

BMC Bioinformatics

Table 4 Classification results using features selected by Student t test.

Algorithm	Accuracy(%)	TP rate	FP rate	TN rate	FN rate	Sensitivity	Specificity	Precision	Fmeasure	RMSE
C4.5	0.6444	0.99	0.79	0.21	0.01	0.99	0.21	0.61	0.76	0.4687
Random Forest	0.6500	0.79	0.53	0.48	0.21	0.79	0.48	0.65	0.71	0.4569
Bagging	0.6833	0.78	0.44	0.56	0.22	0.78	0.56	0.69	0.73	0.4285
Logitboost	0.6889	0.83	0.49	0.51	0.17	0.83	0.51	0.69	0.75	0.4402
Stacking	0.6444	0.99	0.79	0.21	0.01	0.99	0.21	0.61	0.76	0.4761
Adaboost	0.6444	0.77	0.51	0.49	0.23	0.77	0.49	0.69	0.69	0.4412
Multiboost	0.6889	0.81	0.46	0.54	0.19	0.81	0.54	0.70	0.74	0.5175
Logistic	0.7500	0.79	0.30	0.70	0.21	0.79	0.70	0.78	0.78	0.4224
Naivebayes	0.6833	0.64	0.26	0.74	0.36	0.64	0.74	0.76	0.68	0.5289
Bayesnet	0.6722	0.63	0.28	0.73	0.37	0.63	0.73	0.74	0.67	0.5308
Neural Network	0.7000	0.70	0.30	0.70	0.30	0.70	0.70	0.75	0.72	0.4517
RBFnet	0.6722	0.76	0.44	0.56	0.24	0.76	0.56	0.69	0.71	0.4632
SVM	0.6944	0.71	0.33	0.68	0.29	0.71	0.68	0.74	0.71	0.5489

TP rate: True positive rate, FP rate: False positive rate, TN rate: True negative rate, FN rate: False negative rate, RMSE: Root Mean Squared Error. RBFnet: Radio Basis Function network, SVM: Support Vector Machine.

ISSN: 1471-2105