Fishery Knowledge Discovery Based on SVM and Fuzzy Rule Eztraction
In the area of ocean fisheries research, one hotspot is the use of marine environment factors for fisheries forecast. This paper fits in the category using fishery knowledge discovery based on Support Vector Machine (SVM) and fuzzy rule extraction. It takes the Indian Ocean big-eye tuna fishery as its testing ground. Firstly, the support vectors are obtained by training the SVM with some sample data. Then the rules are extracted by the fuzzy classifier method. Meanwhile, a fishery forecasting model is established based on Support Vector Regression (SVR). Experimental results show that the fishery knowledge obtained is of a forceful interpretive capacity, which is ideal for explaining the formation mechanism of fishing grounds. The established fishery forecasting model provides a high level of information accuracy which can be further enhanced by additional fishing effort.
support vector machine fuzzy classification rule eztraction support vector regression Indian Ocean bigeye tuna
Yuan Hong-chun Li Ying Chen Ying
College of Information Technology Shanghai Ocean University Shanghai, China School of Computing & Information Systems University of Tasmania Hobart, Tasmania, Australia
国际会议
北京
英文
828-832
2009-08-08（万方平台首次上网日期，不代表论文的发表时间）