Integrated Evaluation of Candidate Partners Achievement of Dynamic Alliance Based on PCA-SVR

摘要：

Based on the index system of candidate partners achievement of dynamic alliance, a integrated evaluation model is established by using principal component analysis and support vector machine. The method has advantages of accuracy, convenience, reliability and rapidity. Through principal component analysis, we have synthesized numerous criteria, eliminated information overlapping of the samples, and reduced the input dimension of SVM. The method is illustrated through examples, the results obtained by using principal component analysis and support vector machine method are compared with that from neural network method and empirical analysis, and the results show that the conjoint method (PCA-SVR) is more precise and fits better.

关键词： dynamic alliance candidate partners support vector regression principal component analysis achievement evaluation

作者: Yan Qisheng Rao Zhiyong

作者单位: School of Information Jiangnan University, Wuxi 214122, China School of Mathematics & Information Sc School of Mathematics & Information Science East China Institute of Technology Fuzhou 344000, China

会议类型: 国际会议

会议名称: The 2010 International Conference on Computer Application and System Modeling(2010计算机应用与系统建模国际会议 ICCASM 2010)

会议地点: 太原

会议语种:英文

页码: 573-577

在线出版日期: 2010-10-22（万方平台首次上网日期，不代表论文的发表时间）