| SVM决策树在沉积微相识别中的应用 |
| |
| 引用本文： | 展文宁,蔡英,李一民,王玉杰.SVM决策树在沉积微相识别中的应用[J].科技广场,2007(7):144-146. |
| |
| 作者姓名： | 展文宁 蔡英 李一民 王玉杰 |
| |
| 作者单位： | 昆明理工大学信息工程与自动化学院,云南,昆明,650051 |
| |
| 摘 要： | 支持向量机(SVM)作为统计学理论最年轻的分支,其应用日益广泛。针对油层沉积微相的多类识别问题,可采用支持向量机和决策树相结合的方法。对传统的SVM决策树进行改进的基础上,在SVM核函数选取过程中,构造了与实际问题有关的核函数。此方法有效的降低了支持向量机的设计难度,同时提高了识别精度和泛化能力。最后用实例对比神经网络验证了该方法的优越性。
|
| 关 键 词： | 支持向量机 决策树 沉积微相 核函数 识别 |
| 文章编号： | 1671-4792-（2007）7-0113-03 |
the Application of SVM Decision Tree in Sedimentary Facies Recognition |
| |
| Zhan Wenning,Cai Ying,Li Yimin,Wang Yujie.the Application of SVM Decision Tree in Sedimentary Facies Recognition[J].Science Mosaic,2007(7):144-146. |
| |
| Authors: | Zhan Wenning Cai Ying Li Yimin Wang Yujie |
| |
| Institution: | Faculty of Information and Automation, Kunming University of Science and Technology, Yunnan Kunming 650051 |
| |
| Abstract: | As the newest branch of statistic learning theory, support vector machine(SVM) have become increasingly popular tools. Aiming at the recognition of multi-class sedimentary facies in oil layer,the paper presented a method which combine SVM and the decision tree.The traditional SVM decision tree had been improved here ,and during the process of choosing kernel function ,we built a kernel function that can be related to practice. This method effectively reduce the difficulty on designing SVM ,meanwhile raise the recognition rate and generalization ability. The method was tested to be better than neural network by using the example at last. |
| |
| Keywords: | Support Vector Machine Decision Tree Sedimentary Facies Kernel Function Recognition |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
