Standardizing catch per unit effort by machine learning techniques in longline fisheries: a case study of bigeye tuna in the Atlantic Ocean

Yang, Shenglong; Dai, Yang; Fan, Wei; Shi, Huiming

doi:10.1590/S2675-28242020068226

Shenglong Yang

Key Laboratory of Oceanic and Polar Fisheries, Ministry of Agriculture and Rural Affairs; East China Sea Fisheries Research Institute, Chinese Academy of Fishery Sciences, Shanghai, 200090, China

Key and Open Laboratory of Remote Sensing Information Technology in Fishing Resource, East China Sea Fisheries Research Institute, Chinese Academy of Fishery Sciences, Shanghai 200090, China

Yang Dai

Key and Open Laboratory of Remote Sensing Information Technology in Fishing Resource, East China Sea Fisheries Research Institute, Chinese Academy of Fishery Sciences, Shanghai 200090, China

Wei Fan ^**Corresponding author: fanwee@126.com

Key and Open Laboratory of Remote Sensing Information Technology in Fishing Resource, East China Sea Fisheries Research Institute, Chinese Academy of Fishery Sciences, Shanghai 200090, China

http://orcid.org/0000-0001-5036-8733

Huiming Shi

Key Laboratory of Oceanic and Polar Fisheries, Ministry of Agriculture and Rural Affairs; East China Sea Fisheries Research Institute, Chinese Academy of Fishery Sciences, Shanghai, 200090, China

*Corresponding author: fanwee@126.com

Associate Editor: Maria Gasalla

Editor: Rubens M. Lopes

	c	g
Grid search	5.66	64
GA-SVM	5.77	50
PSO-SVM	7.12	60.25

Model	MSE			MAE
Model	Training data	Testing data	All data	Training data	Testing data	All data
Grid search	3.62	8.64*	4.62	0.68	1.81*	0.90*
GA-SVM	0.57	12.37	2.94	0.51	2.11	0.83
PSO-SVM	0.4*	10.92	2.5*	0.46*	2.02	0.77*

Model	Person's			Kendall's rank			Spearman's rank
Model	Training data	Testing data	All data	Training data	Testing data	All data	Training data	Testing data	All data
Grid search	0.88	0.69*	0.85	0.85	0.55*	0.77	0.95	0.73*	0.9
GA-SVM	0.99	0.6	0.91	0.92	0.52	0.83	0.98	0.65	0.92
PSO-SVM	0.99*	0.61	0.92*	0.94*	0.52	0.84*	0.99*	0.67	0.92*

Model	NMSE
Model	Training data	Testing data	All data
Grid search	0.24	0.53*	0.3
GA-SVM	0.04	0.75	0.2
PSO-SVM	0.03*	0.730	0.16*

Variable	d.f.	F	p	Variable	d.f.	F	p
s(year)	3	59.21	<0.000001	s(UTBT)	2.99	64.82	<0.000001
s(month)	2	9.68	0.00032	s(UDBT)	2.86	51.33	<0.000001
s(lat)	3	96.23	< 2e-16	s(LTBT)	2	84.98	<0.000001
s(lon)	3	46.65	<0.000001	s(LDBT)	2.45	66.63	<0.000001

Deviance explained (%)	AIC	R² (adj)	MSE	MAE	Person's	Kendall's rank	Spearman's rank	NMSE
29.6	11381.74	0.293	13.38	2.33	0.44	0.37	0.52	0.86

s . t . \{\begin{matrix} y_{i} - w • φ (x_{i}) - b \leq ε + ξ_{i}^{+} \\ - y_{i} + w • φ (x_{i}) + b \leq ε + ξ_{i}^{-}, i = 1, 2, \dots 1 . \\ ξ_{i}^{+} \geq 0, ξ_{i}^{-} \end{matrix}

[1] *Corresponding author: fanwee@126.com

Brasil

Brasil

Standardizing catch per unit effort by machine learning techniques in longline fisheries: a case study of bigeye tuna in the Atlantic Ocean