支持向量机方法在地球物理学中的应用与展望  

Application and Expectation of Support Vector Machine Method in Geophysics

作　　者：李希亮[1] 李栋梁 张玲[1] 董晓娜[1] 吴丹桐[1] 李霞[1] 金鹏[1] 赵小贺[1] 

机构地区：[1]山东省地震局,山东济南250014 [2]潍坊市地震局,山东潍坊261041

出　　处：《华南地震》2012年第4期52-58,共7页South China Journal of Seismology

摘　　要：支持向量机方法是基于统计学习理论的新一代学习算法,在解决小样本、非线性和高维模式识别中表现出很多特有的优势。通过支持向量机分类的原理,以及在地震勘探、油气储层预测、大地电磁和地震监测中的应用现状进行简要的综述,说明此方法在地球物理学研究中是有效的。随着支持向量机方法不断完善和改进,在地球物理学中的应用前景将更加广泛。Support Vector Machine method is a machine leaning method based on statistical learning theory. It shows obvious advantages in processing small-sample, high dimensional pattern recognition, and nonlinear problems. In this paper, its classification principle and application status in seismic exploration, the oil gas reservoir prediction, magnetotelluric and earthquake monitoring are briefly reviewed, indicating that it is effective in geophysical research. Support vector machine method will be more extensive applied in geophysics as it continuously refines and improves.

关 键 词：支持向量机 地球物理学 展望 

分 类 号：P315.21[天文地球—地震学]