一种基于样本近邻分类精度的支持向量机集成方法
|
摘要
提出了一种基于样本近邻分类精度的支持向量机集成方法.对待分类样本,可通过改进的FCM与模糊贴近度的搜索算法,自动确定其在模糊特征空间集上的有效邻域;在此基础上,依据各分类器在样本近邻的分类精度及设置的阀值,自动选取部分优秀的个体分类器,进行集成判决.实验结果表明,在缩短分类判别时间的情况下,该方法可有效提高分类器性能.
An adaptive support vector machine integrated method was proposed which based on nearest neighbor classification accuracy.For the classified samples,the effective neighborhood on the fuzzy feature space set was determined automatically by using the improved FCM and fuzzy nearness degree search algorithm.Based on classification accuracy and threshold,the model could dynamically select a set of optimal individual classifier to integrate.Experimental results show that the proposed method can improve the classification performance and shorten discrimination time.
An adaptive support vector machine integrated method was proposed which based on nearest neighbor classification accuracy.For the classified samples,the effective neighborhood on the fuzzy feature space set was determined automatically by using the improved FCM and fuzzy nearness degree search algorithm.Based on classification accuracy and threshold,the model could dynamically select a set of optimal individual classifier to integrate.Experimental results show that the proposed method can improve the classification performance and shorten discrimination time.
引文
[1]吕晓燕,罗立民,李祥生.基于支持向量机的冠心病辅助诊断研究[J].计算机工程与应用,2008,44(36):221-225.LüXiaoyan,Luo Liming,Li Xiangsheng.Studies on appilcation of support vector machine in detection of coronary heart disease[J].Computer Engineering and Applications,2009,44(36):221-225.(in Chinese)
[2]史荧中,曹建峰,邓赵红.面向数据流分类的柔性漂移支持向量机[J].计算机工程与应用,2017,53(23):118-122.Shi Yingzhong,Cao Jianfeng,Deng Zhaohong.Flexible drift support vector machines for data stream classification[J].Computer Engineering and Applications,2017,53(23):118-122.(in Chinese)
[3]Zhou Z H,Wu J,Tang W.Ensembling neural networks:Many could be better than al1[J].Artificial Intelligence,2002,137(1-2):239-263.
[4]Kim K I,Jung K,Park S H.Support vector machines for texture classification[J].IEEE Trans Patern Analysis&Machine Intelligence, 2002, 24(11):1542-1550.
[5] Valentini G,Dietterich T.Bias-variance analysis of support vector machines for the development of svm based ensemble methods[J].Journal of Machine Learning Research,2014,50(3):75-78.
[6]He Jieyue,Rick Chow,Zhong Wei.Clinical charge profiles prediction for patients diagnosed with chronic diseases using multi-level support vector machine[J].Expert Systems with Application,2012,39(1):1474-1483.
[7]周华平,李敬兆.多阶灰色支持向量机集成预测模型研究[J].计算机工程与科学,2015,37(3):539-546.Zhou Huaping,Li Jingzhao.An integrated prediction model using multi-stage gray model and support vector machine[J]. Computer Engineering and Science,2015,37(3):539-546.(in Chinese)
[8]米爱中,陆瑶.基于聚类与排序修剪的分类器集成方法[J].计算机应用研究,2018,35(7):2034-2037.Mi Aizhong,Lu Yao.Classifier ensemble method based on clustering and sorting pruning[J].Application Research of Computers,2018,35(7):2034-2037.(in Chinese)
[9]李明峰,贾修一.基于多分类器集成学习的中文反语识别技术[J].计算机与数字工程,2018,46(9):1790-1795.Li Mingfeng,Jia Xiuyi.Chinese irony recognition based on ensemble learning of multiple classifiers[J].Computer and Digital Engineering,2018,46(9):1790-1795.(in Chinese)
[10]张文博,姬红兵,王磊.一种自适应权值的多特征融合分类方法[J].系统工程与电子技术,2013,35(6):1133-1137.Zhang Wenbo,Ji Hongbing,Wang Lei.Adaptive weighted feature fusion classification method[J].Systems Engineering and Electronics,2013,35(6):1133-1137.(in Chinese)
[2]史荧中,曹建峰,邓赵红.面向数据流分类的柔性漂移支持向量机[J].计算机工程与应用,2017,53(23):118-122.Shi Yingzhong,Cao Jianfeng,Deng Zhaohong.Flexible drift support vector machines for data stream classification[J].Computer Engineering and Applications,2017,53(23):118-122.(in Chinese)
[3]Zhou Z H,Wu J,Tang W.Ensembling neural networks:Many could be better than al1[J].Artificial Intelligence,2002,137(1-2):239-263.
[4]Kim K I,Jung K,Park S H.Support vector machines for texture classification[J].IEEE Trans Patern Analysis&Machine Intelligence, 2002, 24(11):1542-1550.
[5] Valentini G,Dietterich T.Bias-variance analysis of support vector machines for the development of svm based ensemble methods[J].Journal of Machine Learning Research,2014,50(3):75-78.
[6]He Jieyue,Rick Chow,Zhong Wei.Clinical charge profiles prediction for patients diagnosed with chronic diseases using multi-level support vector machine[J].Expert Systems with Application,2012,39(1):1474-1483.
[7]周华平,李敬兆.多阶灰色支持向量机集成预测模型研究[J].计算机工程与科学,2015,37(3):539-546.Zhou Huaping,Li Jingzhao.An integrated prediction model using multi-stage gray model and support vector machine[J]. Computer Engineering and Science,2015,37(3):539-546.(in Chinese)
[8]米爱中,陆瑶.基于聚类与排序修剪的分类器集成方法[J].计算机应用研究,2018,35(7):2034-2037.Mi Aizhong,Lu Yao.Classifier ensemble method based on clustering and sorting pruning[J].Application Research of Computers,2018,35(7):2034-2037.(in Chinese)
[9]李明峰,贾修一.基于多分类器集成学习的中文反语识别技术[J].计算机与数字工程,2018,46(9):1790-1795.Li Mingfeng,Jia Xiuyi.Chinese irony recognition based on ensemble learning of multiple classifiers[J].Computer and Digital Engineering,2018,46(9):1790-1795.(in Chinese)
[10]张文博,姬红兵,王磊.一种自适应权值的多特征融合分类方法[J].系统工程与电子技术,2013,35(6):1133-1137.Zhang Wenbo,Ji Hongbing,Wang Lei.Adaptive weighted feature fusion classification method[J].Systems Engineering and Electronics,2013,35(6):1133-1137.(in Chinese)