| 基于DAGSVM的装备不常用备件分类方法 |
| |
| 引用本文： | 姜晨,徐廷学,余仁波.基于DAGSVM的装备不常用备件分类方法[J].舰船科学技术,2011,33(7). |
| |
| 作者姓名： | 姜晨 徐廷学 余仁波 |
| |
| 作者单位： | 1. 中国人民解放军91872部队,北京,102442
2. 海军航空工程学院,山东烟台,264001 |
| |
| 摘 要： | 针对装备不常用备件需求样本数据有限、影响不常用备件需求的因素众多,且各因素间的关系多为非线性等特点,分析了现有备件分类方法中的不足,并引入在小样本学习方面具有优势的支持向量机(Support Vector Machines,SVM)方法,利用有向无环图支持向量机(Directed Acyclic Graph SVM,DAGSVM)方法对装备的不常用备件进行分类,并进行了实例应用。结果表明,该方法可以有效地解决装备不常用备件的多类分类问题。
|
| 关 键 词： | 不常用备件 SVM方法 DAGSVM方法 |
Classified method equipment rarely used spare parts based on DAGSVM |
| |
| JIANG Chen,XU Ting-xue,YU Ren-bo.Classified method equipment rarely used spare parts based on DAGSVM[J].Ship Science and Technology,2011,33(7). |
| |
| Authors: | JIANG Chen XU Ting-xue YU Ren-bo |
| |
| Institution: | JIANG Chen1,XU Ting-xue2,YU Ren-bo2(1.Unit No.91872 of PLA,Beijing102442,China,2.Naval Aeronautical and Astronautical University,Yantai264001,China) |
| |
| Abstract: | Aiming at the characters that the simple data of equipment rarely used spare parts demand allocation is limited,the influence factors of spare parts demand allocation are numerous,and the relations between which are almost nonlinear,this paper analyze shortcomings about the present spare parts classificaiton methods,and induce the method of SVM which preponderant in little simple learning.This paper using DAGSVM arithmetic to classify equipment rarely used spare parts,and give an application example.The res... |
| |
| Keywords: | rarely used spare parts SVM method DAGSVM arithmetic |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
|
