计算机科学 ›› 2016, Vol. 43 ›› Issue (6): 276-279.doi: 10.11896/j.issn.1002-137X.2016.06.054
范旭慧,张捷,班登科
FAN Xu-hui, ZHANG Jie and BAN Deng-ke
摘要： 为了解决半监督支持向量分类优化模型中的非凸非光滑问题,基于分段逼近的思想提出了一个分段函数,并以此逼近非凸非光滑的目标函数。给出的分段函数可以根据不同的精度要求选择不同的逼近参数,同时构造出基于上述分段函数的光滑半监督支持向量机模型。采用了LDS(Low Density Separation)算法求解模型,分析了其对对称铰链损失函数的逼进精度。理论分析和数值实验结果都证明分段光滑的半监督支持向量机的分类性能和效率优于以往提出的光滑模型。
| [1] Burges C.A tutorial on support vector machines for pattern re-cognition[J].Data Mining and Knowledge Discovery,1998,2(2):127-167 [2] Lu S X,Tian R N.Structural Weighted Least Squares Support Vector Machine Classifier[J].Computer Science,2013,40(12):52-54(in Chinese) 鲁淑霞,田如娜.结构化加权最小二乘支持向量机[J].计算机科学,2013,40(12):52-54 [3] Sun L,Jing L,Xia X D.A new Proximal Support vector Machine for Semi-supervised Classification[M].Heidelberg:Springer Berlin,2006:23-34 [4] Bennett K P,Demiriz A.Semi-supervised support vector ma-chines[M].Cambridge:MIT Press,1998:10-20 [5] Zhang X D,Ma J.A general cubic spline smooth semi-supervised support vector machine[J].Chinese Journal of Engineering,2015,37(3):386-389(in Chinese) 张晓丹,马菁.一个广义三次样条光滑半监督支持向量机[J].工程科学学报,2015,37(3):386-389 [6] Chapelle O,Zien A.Semi-supervised classification by low density separation[C]∥10th International Workshop on Artificial Intelligence and Statistics.Barbados,2005:57-68 [7] Liu Y Q,Lin S Y.Poliynomial Smooth Semi-supervised Support Vector Classifier[J].Computer Science,2009,36(7):179-181(in Chinese) 刘叶青,刘三阳.一种多项式光滑的半监督支持向量机[J].计算机科学,2009,36(7):179-181 [8] Bennett K,Demiriz A.Semi-supervised support vector machines[C]∥Proceedings of the 1998 Conference Advances in Neural Information Processing Systems.1998:368-374 [9] Joachims T.Transductive inference for text classication using support vector machines[C]∥Internaltional Conference on Machine Learning,Pittsburgh.1999:200-209 [10] Dijkstra E W.A note on two problems in connection with graphs[J].Numerische Mathematics,1959,1(1):269-271 [11] Cox T F,Cox M A.Multidimensional Scaling[J].Journal of the Royal Statistical Society Series A(Statistics in Society),1994,5(2):271-281 [12] Bengio Y,Grandvalet Y.Semi-supervised learning by entropy minimization[J].Neural Information Processing Systems,2005,5(2):529-536 [13] Nene S A,Nayar S K,Murase H.Columbia object image library (coil-20)[R].CUCS-005-96,Columbia Univ.,USA,February 1996 |
| No related articles found! |
|
||
首页