Trend prediction of the 3D thermocline’s lateral boundary based on the SVR method

EURASIP Journal on Wireless Communications and Networking

Table 1 The evaluation for prediction results

Year-Mon	Tp	Ture	Predict	Precision	Recall	F1
1601	122	200	156	0.61	0.782051282	0.685393258
1602	124	167	142	0.74251497	0.873239437	0.802588997
1603	54	132	71	0.409090909	0.76056338	0.532019704
1604	54	153	65	0.352941176	0.830769231	0.495412844
1605	39	127	42	0.307086614	0.928571429	0.461538462
1606	66	117	84	0.564102564	0.785714286	0.656716418
1607	81	122	110	0.663934426	0.736363636	0.698275862
1608	87	165	107	0.527272727	0.813084112	0.639705882
1609	81	159	98	0.509433962	0.826530612	0.630350195
1610	95	176	118	0.539772727	0.805084746	0.646258503
1611	133	215	159	0.618604651	0.836477987	0.711229947
1612	121	208	134	0.581730769	0.902985075	0.707602339