Table 3.

Cross-validation for common peaks.

	Training sample size (%)	TP	TN	FP	FN	Sensitivity	Specificity	Accuracy
Binary logistic regression	0.4	80169	10970	3030	7831	0.9110	0.7836	0.8935
	0.6	52459	7542	1458	5541	0.9045	0.8380	0.8955
	0.8	26258	4256	744	2742	0.9054	0.8512	0.8975
SVM	0.4	81890	9144	4856	6110	0.9306	0.6531	0.8925
	0.6	54293	6543	2457	3707	0.9361	0.7270	0.9080
	0.8	27333	3769	1231	1667	0.9425	0.7538	0.9148
LDA	0.4	81219	11258	2742	6781	0.9229	0.8041	0.9066
	0.6	54271	7380	1620	3729	0.9357	0.8200	0.9202
	0.8	27320	4021	979	1680	0.9421	0.8042	0.9218
QDA	0.4	78654	8663	5337	9346	0.8938	0.6188	0.8560
	0.6	50593	6991	2009	7407	0.8723	0.7768	0.8595
	0.8	25080	4048	952	3920	0.8648	0.8096	0.8567
Neura Networks	0.4	86002	9173	5197	1998	0.9773	0.6383	0.9297
	0.6	56730	6180	2820	1270	0.9781	0.6867	0.9390
	0.8	28511	3341	1659	489	0.9831	0.6682	0.9368
Classification Trees	0.4	80836	3530	10470	7164	0.9186	0.2521	0.8271
	0.6	53349	2623	6377	4651	0.9198	0.2914	0.8354
	0.8	26795	1439	3561	2205	0.9240	0.2878	0.8304
Boosting Trees	0.4	79569	2546	9603	1164	0.9856	0.2096	0.8841
	0.6	52099	1478	8392	852	0.9839	0.1497	0.8529
	0.8	25924	708	9136	440	0.9833	0.0719	0.7355