TABLE 2.

Accuracy of MAPIC and competitors. Among shapelet-based classifiers, bold indicates the best performer and italic the second one (the higher the better).

Dataset	MAPIC	mpdt	opt	bf	dtc	knn
ArrowHead	0.53	0.39	0.39	0.39	0.39	0.64
BirdChicken	0.80	0.50	0.50	0.50	0.50	0.45
Coffee	0.96	0.54	0.54	0.54	0.54	0.89
Earthquakes	0.74	0.73	0.75	0.71	0.68	0.74
ECG5000	0.91	0.46	0.91	0.90	0.90	0.87
ECG200	0.78	0.64	0.68	0.78	0.76	0.85
FaceFour	0.24	0.16	0.16	0.16	0.16	0.68
GunPoint	0.91	0.73	0.71	0.75	0.63	0.85
Phalanges	0.65	0.62	0.67	0.68	0.69	0.66
PowerDemand	0.92	0.53	0.61	0.94	0.97	0.93
Strawberry	0.87	0.64	0.85	0.90	0.88	0.94
Trace	0.98	0.66	0.89	0.98	0.72	0.74
TwoLeadECG	0.77	0.50	0.50	0.50	0.50	0.58
Wafer	0.95	0.89	0.92	0.95	0.97	0.99
Wine	0.61	0.50	0.74	0.57	0.74	0.72
WormsTwoClass	0.68	0.43	0.60	0.71	0.52	0.61
Yoga	0.66	0.52	0.61	0.62	0.60	0.79
Avg	0.76	0.55	0.64	0.68	0.65	0.76
Rank	2.38	5.26	3.79	3.26	3.79	2.50