DocumentCode
1912046
Title
Research of Coastal Area Ecological Economic System Based on Stochastic Gradient Regression
Author
Jialiang, Guo ; Hongli, Wang
Author_Institution
Sch. of Manage., Tianjin Univ., Tianjin, China
Volume
4
fYear
2009
fDate
10-11 Oct. 2009
Firstpage
322
Lastpage
324
Abstract
Coastal area was defined as the conjunction of land and sea and it had very important ecological, economic and social value. Because the economy developed fast and demand of resource increased heavily, there were serious pressures on the ecological economic system of coastal area, such as large amount land-sourced pollutant and unreasonable fishing. The gross output value of fishery at Tianjin, which is a seaside city of Pohai, was studied in order to analyze the influence on the ecological economic system of coastal area from those stress factors. Prediction model of the gross output value of fishery at Tianjin was established by stochastic gradient regression algorithm. The influence on the system from those stress factors was analyzed by the model. Comparing to the prediction results of models using support vector machine method and adaptive spline regression model, it was concluded that stochastic gradient regression method performed better.
Keywords
environmental economics; regression analysis; support vector machines; adaptive spline regression model; coastal area; ecological economic system; stochastic gradient regression; support vector machine method; Aquaculture; Biological system modeling; Cities and towns; Economic forecasting; Land pollution; Predictive models; Sea measurements; Stochastic processes; Stochastic systems; Stress; ecological economic system of coastal area; gradient boosting; prediction; stochastic gradient regression model;
fLanguage
English
Publisher
ieee
Conference_Titel
Intelligent Computation Technology and Automation, 2009. ICICTA '09. Second International Conference on
Conference_Location
Changsha, Hunan
Print_ISBN
978-0-7695-3804-4
Type
conf
DOI
10.1109/ICICTA.2009.793
Filename
5288286
Link To Document