Table 4.

Summary of results using the shape-based proposal with different classifiers.

Pattern Selection Method	Classifier	No. of Patterns	TP	FP	TN	FN	Sensitivity	Specificity	Accuracy
RBF SVM support vectors	RBF SVM	1551	4724	33	486	221	0.9553	0.9364	0.9535
		221	4611	121	398	334	0.9325	0.7669	0.9167
		5	4586	126	393	359	0.9274	0.7572	0.9112
	Random Forests	1551	4674	35	484	271	0.9452	0.9326	0.9440
		221	4573	75	444	372	0.9248	0.8555	0.9182
		5	4378	109	410	567	0.8853	0.7900	0.8763
	GBM	1551	4668	37	482	277	0.9440	0.9287	0.9425
		221	4547	82	437	398	0.9195	0.8420	0.9122
		5	4497	116	403	448	0.9094	0.7765	0.8968
	Linear SVM	1551	4621	37	482	324	0.9345	0.9287	0.9339
		221	4449	52	467	496	0.8997	0.8998	0.8997
		5	4425	115	404	520	0.8948	0.7784	0.8838
	kNN ($k=7$)	1551	4703	69	450	242	0.9511	0.8671	0.9431
		221	4563	68	41	382	0.9228	0.8690	0.9176
		5	4330	105	414	615	0.8756	0.7977	0.8682
	Naïve Bayes	1551	4660	146	373	285	0.9424	0.7187	0.9211
		221	4607	132	387	338	0.9316	0.7457	0.9140
		5	4541	123	396	404	0.9183	0.7630	0.9036
	C5.0	1551	4400	56	463	545	0.8898	0.8921	0.8900
		221	4176	84	435	769	0.8445	0.8382	0.9439
		5	4683	136	383	262	0.9470	0.7380	0.9272
PAM medoids	RBF SVM	180	4651	47	472	294	0.9405	0.9094	0.9376
		10	4555	81	438	390	0.9211	0.8439	0.9138
		4	4623	104	415	322	0.9349	0.7996	0.9220
		2	4323	149	370	622	0.8742	0.7129	0.8589
	Random Forests	180	4633	57	462	312	0.9369	0.8902	0.9325
		10	4513	77	442	432	0.9126	0.8516	0.9068
		4	4410	91	428	535	0.8918	0.8247	0.8854
		2	3973	126	393	972	0.8034	0.7572	0.7990
	GBM	180	4598	53	466	347	0.9298	0.8979	0.9268
		10	4468	70	449	447	0.9035	0.8651	0.8999
		4	4500	94	425	445	0.9100	0.8189	0.9014
		2	4229	120	399	716	0.8552	0.7688	0.8470
	Linear SVM	180	4511	38	481	434	0.9122	0.9268	0.9136
		10	4544	89	430	401	0.9189	0.8285	0.9103
		4	4496	123	396	449	0.9092	0.7630	0.8953
		2	4311	156	363	634	0.8718	0.6994	0.8554
	kNN ($k=7$)	180	4629	66	453	316	0.9361	0.8728	0.9301
		10	4572	97	422	373	0.9246	0.8131	0.9140
		4	4434	92	425	445	0.9100	0.8189	0.9014
		2	4117	120	399	828	0.8326	0.7688	0.8265
	Naïve Bayes	180	4526	113	406	419	0.9153	0.7823	0.9026
		10	4346	79	440	599	0.8789	0.8478	0.8759
		4	4395	85	434	550	0.8888	0.8362	0.8838
		2	4172	156	363	773	0.8437	0.6994	0.8300
	C5.0	180	4362	76	443	583	0.8821	0.8536	0.8794
		10	4293	77	442	652	0.8681	0.8516	0.8666
		4	4593	109	410	352	0.9288	0.7900	0.9156
		2	4200	144	375	745	0.8493	0.7225	0.8373
Exhaustive search	RBF SVM	2	4492	93	426	453	0.9084	0.8208	0.9001
	Random Forests	2	4179	89	430	766	0.8451	0.8285	0.8435
	GBM	2	4306	78	441	639	0.8708	0.8497	0.8688
	Linear SVM	2	4360	85	434	585	0.8817	0.8362	0.8774
	kNN ($k=7$)	2	4293	91	428	625	0.8681	0.8247	0.8640
	Naïve Bayes	2	4135	91	428	810	0.8362	0.8247	0.8351
	C5.0	2	4587	120	399	358	0.9276	0.7688	0.9125
Informed search: Breadth-first search	RBF SVM	4	4526	68	451	419	0.9153	0.8690	0.9109
		10	4504	63	456	441	0.9108	0.8786	0.9078
	Random Forests	4	4441	68	451	504	0.8981	0.8690	0.8953
		10	4494	60	459	451	0.9088	0.8844	0.9065
	GBM	4	4411	62	457	534	0.8920	0.8805	0.8909
		10	4465	64	455	480	0.9029	0.8767	0.9004
	Linear SVM	4	4376	66	453	569	0.8849	0.8728	0.8838
		10	4443	69	450	502	0.8985	0.8671	0.8955
	kNN ($k=7$)	4	4434	75	444	511	0.8967	0.8555	0.8928
		10	4430	74	445	515	0.8959	0.8574	0.8922
	Naïve Bayes	4	4623	130	389	322	0.9349	0.7495	0.9173
		10	4645	110	409	300	0.9393	0.7881	0.9250
	C5.0	4	4404	86	433	541	0.8906	0.8343	0.8852
		10	4382	64	455	563	0.8861	0.8767	0.8852
Informed search: Simulated Annealing	RBF SVM	4	4656	121	398	289	0.9416	0.7669	0.9250
		10	4532	92	427	413	0.9165	0.8227	0.9076
	Random Forests	4	4414	95	424	531	0.8926	0.8170	0.8854
		10	4496	72	447	449	0.9092	0.8613	0.9046
	GBM	4	4524	110	409	421	0.9128	0.7881	0.9028
		10	4441	83	436	504	0.8981	0.8401	0.8926
	Linear SVM	4	4553	124	395	392	0.9207	0.7611	0.9056
		10	4384	85	434	561	0.8866	0.8362	0.8818
	kNN ($k=7$)	4	4443	114	405	502	0.8985	0.7803	0.8873
		10	4471	98	421	747	0.9041	0.8112	0.8953
	Naïve Bayes	4	4546	142	377	399	0.9193	0.7264	0.9010
		10	4595	135	384	350	0.9292	0.7399	0.9112
	C5.0	4	4413	101	418	535	0.8924	0.8054	0.8842
		10	4134	71	448	811	0.8360	0.8632	0.8386