|
|||||
|
|
| 一种新的模糊支持向量回归方法 | |
| 引用本文： | 赵时,马曾,嵇艳.一种新的模糊支持向量回归方法[J].青岛大学学报(自然科学版),2010,23(3):47-51. |
| 作者姓名： | 赵时 马曾 嵇艳 |
| 作者单位： | 1. 海军航空工程学院青岛分院,山东,青岛,266041 2. 海军潜艇学院,山东,青岛,266071 3. 92995部队,山东,青岛,266041 |
| 摘 要： | 为克服传统的模糊支持向量机隶属度函数都是基于样本与类中心距离进行设计所带来的局限性问题,提出了基于样本到超平面距离的新隶属度函数设计方法。该方法从支持向量机的回归本质出发,通过更加合理地设计隶属度函数,提高支持向量机的回归的泛化鲁棒能力。仿真结果证明,该方法具有更好的鲁棒性,提高了模糊支持向量机的泛化能力。 |
| 关 键 词： | 支持向量机 模糊支持向量机 隶属度函数 |
A New Fuzzy Support Vector Regression Method |
|
| ZHAO Shi,MA Zeng,JI Yan.A New Fuzzy Support Vector Regression Method[J].Journal of Qingdao University(Natural Science Edition),2010,23(3):47-51. | |
| Authors: | ZHAO Shi MA Zeng JI Yan |
| Institution: | 1.Qingdao Branch,Naval Aeronautical Engineering Institute,Qingdao 266041,China;2.Navy Submarine Academy,Qingdao 266071,China;3.Unit of PLA 92995,Qingdao 266041,China) |
| Abstract: | To overcome the limitation that the membership function for fuzzy support vector machines is designed with the distance between sample and center,a new designed method for membership function is proposed based on the distance of sample from planoid.The method support vector machines in returning essence,and the return pan-robust ability of vector machines is improved through reasonable design.Simulation results show that this methed has a better rabustuess,and inprove the generatization ability of fuzzy support vector. |
| Keywords: | support vector machine fuzzy support vector machine membership function |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ | |