Machine-readable expression of unstructured geological information and intelligent prediction of mineralization associated anomaly areas in Pangxidong District, Guangdong, China
本研究中应用了两种异常检测算法。第一种算法为单分类支持向量机(One-Class Support Vector Machine, OCSVM),它是一种无监督分类算法,经常用于解决异常检测和极度不平衡的数据挖掘两方面的问题[14-15]。与经典支持向量机的不同之处在于,单分类支持向量机属于无监督学习,并且其所要分类的类别只有一类。
由于是无监督学习,因此单分类支持向量机不需要对训练样本标记输入标签的[16]。由于只有一类数据的特征信息,与传统支持向量机寻找超平面对数据进行分类的思路不同,它采用支持向量数据描述(Support Vector Data Description, SVDD)。图2[17]展示了SVDD的二维原理。SVDD本质上属于基于边界数据特征的描述方法,其基本思想和理论基础源于支持向量机的经典理论[18-19]。
DAVYM, GODSILLS. Detection of abrupt spectral changes using support vector machines: an application to audio signal segmentation[C]// Proceedings of IEEE international conference on acoustics speech and signal processing. Orlando: IEEE, 2002: 1313-1316.
[15]
LECOMTES, LENGELLER, RICHARDC, et al. Abnormal events detection using unsupervised One-Class SVM - Application to audio surveillance and evaluation[C]// Proceedings of 2011 8th IEEE international conference on advanced video and signal based surveillance (AVSS). Klagenfurt: IEEE, 2011: 124-129.
[16]
SCHÖLKOPFB, PLATTJ C, SHAWE-TAYLORJ C, et al. Estimating the support of a high-dimensional distribution[J]. Neural Computation, 2001, 13(7): 1443-1471.
TAXD M J, DUINR P W. Support vector data description[J]. Machine Learning, 2004, 54(1): 45-66.
[19]
VAPNIKV N. The nature of statistical learning theory[M]. New York: Springer, 1995.
[20]
RUMELHARTD E, HINTONG E, WILLIAMSR J. Learning representations by back-propagating errors[J]. Nature, 1986, 323: 533-536.
[21]
ANJ, CHOS. Variational autoencoder based anomaly detection using reconstruction probability[J]. Special Lecture on IE, 2015, 2(1): 1-18.
[22]
VALENTINEA P, TRAMPERTJ. Data space reduction, quality assessment and searching of seismograms: autoencoder networks for waveform data[J]. Geophysical Journal International, 2012, 189(2): 1183-1202.
[23]
RIFAIS, VINCENTP, MULLERX, et al. Contractive auto-encoders: explicit invariance during feature extraction[C]// Proceedings of the 28th international conference on machine learning. Bellevue: ACM, 2011: 833-840.
CHENGQ M, BONHAM-CARTERG, WANGW L, et al. A spatially weighted principal component analysis for multi-element geochemical data for mapping locations of felsic intrusions in the Gejiu mineral district of Yunnan, China[J]. Computers and Geosciences, 2011, 37(5): 662-669.
[29]
XIAOF, CHENJ G, ZHANGZ Y, et al. Singularity mapping and spatially weighted principal component analysis to identify geochemical anomalies associated with Ag and Pb-Zn polymetallic mineralization in Northwest Zhejiang, China[J]. Journal of Geochemical Exploration, 2012, 122: 90-100.