Synthetic Evaluation Method Based Support Vector Classifier and Regression Machine
-
摘要: 采用支持向量多值分类机和回归机进行综合评价排序,以提高机器学习方法的综合评价排序能力,并以管理信息系统综合评价为例,与人工神经网络(ANN)方法进行了对比研究.试验结果表明,基于支持向量多值分类机综合评价得分之间的差异比ANN更明显,而且基于支持向量回归机综合评价得分的相对误差明显小于ANN.Abstract: To improve the synthetic evaluation and ranking abilities of machine learning methods,a support vector multi-classifier and a support vector regression machine were applied to the ranking of synthetic evaluation.By taking the evaluation of a management information system as an example,a contrast research between the proposed approach and ANN(artificial neural network) was made.Experimental results show that the difference of synthetic evaluation scores obtained by a support vector multi-classifier is more remarkable than that by ANN,and the relative error of synthetic evaluation scores based on a support vector regression machine is smaller than that based on ANN.
-
赵冬梅,张炜.工业企业技术开发水平综合评价[J].西南交通大学学报,1998,33(2):224-228.ZHAO Dongmei,ZHANG Wei.A comprehensive evaluation of technological developments of industrial enterprises[J].Journal of Southwest Jiaotong University,1998,33(2):224-228.[2] 林杰,郭耀煌.用神经网络方法预测股票短期走势[J].西南交通大学学报,1998,33(3):299-305.LIN Jie,GUO Yaohuang.Short term prediction of stock prices based on neural networks[J].Journal of Southwest Jiaotong University,1998,33(3):299-305.[3] 王建成,高大启,王静,等.改进的遗传和BP杂交算法及神经网络经济预警系统设计[J].系统工程理论与实践,1998(4):136-141.WANG Jiancheng,GAO Daqi,WANG Jing,et al.Designing of ANN economic early warning system based on improved genetic and BP algorithms[J].1998(4):136-141.[4] 何正友,钱清泉.小波神经网络改进结构及其学习算法[J].西南交通大学学报,1999,34(4):436-440.HE Zhengyou,QIAN Qingquan.An improved wavelet neural network structure and its learning algorithm[J].Journal of Southwest Jiaotong University,1999,34 (4):436-440.[5] 向小东,郭耀煌,刁尚敏.BP算法的改进及其在股票价格预测中的应用[J].西南交通大学学,2001,36(4):425-427.XIANG Xiaodong,GUO Yaohuang,DIAO Shangmin.Improved BP algorithm and its application in prediction of stock price[J].Journal of Southwest JiaoTong University,2001,36 (4):425-427.[6] 张新红,郑丕谔.基于神经网络的管理信息系统综合评价方法[J].系统工程学报,2002,17(5):445-450.ZHANG Xinhong,ZHENG Pi' e.Neural networks based integrated evaluation method for MIS[J].Journal of Systems Engineering,2002,17(5):445-450.[7] 孙修东,李宗斌,陈富民.基于人工神经网络的多指标综合评价方法研究[J].郑州轻工业学院学报(自然科学版),2003,18(2):11-14.SUN Xiudong,LI Zongbin,CHEN Fumin.Research on multiple attribute synthetical evaluation methods based on artificial neural network[J].Journal of Zhengzhou Institute of Light Industry (Natural Science),2003,18 (2):11-14.[8] 肖健华,吴今培,杨叔子.基于SVM的综合评价方法研究[J].计算机工程,2002,28(8):28-30.XIAO Jianhua,WU Jinpei,YANG Shuzi.Approach of evaluation system based on support vector machine[J].Computer Engineering,2002,28(8):28-30.[9] 范昕炜.支持向量机的算法及其应用[D].博士学位论文,杭州:浙江大学,2003.FAN Xinwei.Support vector machine and its applications[D].Ph.D.Thesis,Dissertation Submitted to Zhejiang University for the Degree of Doctor of Engineering,2003.[10] VAPNIK V.统计学习理论的本质[M].张学工译.北京:清华大学出版社,2000.100-350.[11] VAPNIK V.Statistical learning theory[M].New York John WileySons,1998.80-310.[12] VAPNIK V.An overview of statistical learning theory[J].IEEETrans.on NN.,1999,10 (3):988-999.[13] 邓乃扬,田英杰.数据挖掘中的新方法-支持向量机[M].北京:科学出版社,2004.214-221.[14] 胡永宏,贺思辉.综合评价方法[M].北京:科学出版社,2000.167-182.[15] 张学工.关于统计学习理论与支持向量机[J].自动化学报,2000,26(1):32-42.ZHANG Xuegong.Introduction to statistical learning theory and support vector machines[J].ACTA AUTOMATICA SINICA,2000,26 (1):32-42.[16] PONTIL M,RIFKIN R,EVENIOU T.From regression to classification in support vector machines[C]// Proceedings ESANN.Brussels:D Facto,1999.225-230.[17] SMOLA A,SCHOLKOPF B.On a kernel-based method for pattern,regression,approximation and operator iversion[J].Algorithmica,1998(22):211-231.
点击查看大图
计量
- 文章访问数: 1736
- HTML全文浏览量: 102
- PDF下载量: 329
- 被引次数: 0