Table 6.

The comparison of negative predicted value (NPV) between CA-125 alone and all four proteins (CA-125, MIF, OPN, and PROL) when using four classification techniques, namely K-Nearest Neighbor (KNN), Logistic Regression (LR), Random Forest (RF), and Support Vector Machine (SVM), in various experiments. The NPV measures the liability of a negative test result and is defined as the ratio of the number of control samples correctly identified to the total number of negative test results. We highlighted the cell(s) with best NPV in each experiment. In general, using random forest and all four proteins yield the best results on most occasions.

ID	Training	Testing	CA-125					4 Proteins
			KNN	LR	RF	SVM	Mean	KNN	LR	RF	SVM	Mean
1	E1 vs. H1	E1 vs. H1	0.94	0.97	0.93	0.94	0.95	0.97	0.91	1	0.94	0.96
2	E1 vs. H1	E2 vs. H2	0.96	0.96	0.95	0.95	0.96	0.97	0.96	0.97	0.97	0.97
3	E1 vs. H1	C2 vs. H2	0.84	0.85	0.85	0.83	0.84	0.83	0.81	0.87	0.88	0.85
4	L1 vs. H1	L1 vs. H1	0.86	0.94	0.94	0.86	0.90	1	1	1	1	1.00
5	L1 vs. H1	L2 vs. H2	0.89	0.89	0.89	0.89	0.89	0.91	0.92	0.93	0.92	0.92
6	L1 vs. H1	C2 vs. H2	0.86	0.86	0.86	0.85	0.86	0.89	0.91	0.91	0.91	0.91
7	C1 vs. H1	E2 vs. H2	0.94	0.95	0.94	0.95	0.95	0.94	0.95	0.98	0.96	0.96
8	C1 vs. H1	L2 vs. H2	0.91	0.90	0.89	0.89	0.90	0.91	0.92	0.93	0.92	0.92
9	C1 vs. H1	C2 vs. H2	0.89	0.88	0.86	0.86	0.87	0.89	0.91	0.91	0.91	0.91
10	E2 vs. H2	E2 vs. H2	0.93	0.93	0.93	0.93	0.93	0.94	0.96	0.95	0.94	0.95
11	E2 vs. H2	E1 vs. H1	0.86	0.85	0.86	0.85	0.86	0.93	0.96	0.98	0.96	0.96
12	E2 vs. H2	C1 vs. H1	0.64	0.63	0.65	0.63	0.64	0.62	0.70	0.72	0.78	0.71
13	L2 vs. H2	L2 vs. H2	0.88	0.86	0.87	0.94	0.89	0.91	0.91	0.92	0.91	0.91
14	L2 vs. H2	L1 vs. H1	0.87	0.86	0.84	0.86	0.86	0.93	0.92	0.96	0.92	0.93
15	L2 vs. H2	C1 vs. H1	0.80	0.78	0.77	0.78	0.78	0.90	0.85	0.93	0.89	0.89
16	C2 vs. H2	E1 vs. H1	0.93	0.92	0.92	0.91	0.92	0.99	0.99	0.98	0.99	0.99
17	C2 vs. H2	L1 vs. H1	0.86	0.86	0.85	0.85	0.86	0.92	0.92	0.95	0.92	0.93
18	C2 vs. H2	C1 vs. H1	0.78	0.78	0.80	0.77	0.78	0.91	0.92	0.92	0.90	0.91