Recognition of geochemical anomalies using a deep autoencoder network

被引:226
作者
Xiong, Yihui [1 ]
Zuo, Renguang [1 ]
机构
[1] China Univ Geosci, State Key Lab Geol Proc & Mineral Resources, Wuhan 430074, Peoples R China
基金
中国国家自然科学基金;
关键词
Autoencoder network; Multivariate geochemical data; Geochemical exploration; Skarn-type iron deposits; UNDISCOVERED MINERAL-DEPOSITS; RESTRICTED BOLTZMANN MACHINE; MAKENG FE DEPOSIT; EXPLORATION GEOCHEMISTRY; LITHOGEOCHEMICAL DATA; YUNNAN-PROVINCE; FUJIAN PROVINCE; NEURAL-NETWORKS; COVERED AREAS; CHINA;
D O I
10.1016/j.cageo.2015.10.006
中图分类号
TP39 [计算机的应用];
学科分类号
081203 ; 0835 ;
摘要
In this paper, we train an autoencoder network to encode and reconstruct a geochemical sample population with unknown complex multivariate probability distributions. During the training, small probability samples contribute little to the autoencoder network. These samples can be recognized by the trained model as anomalous samples due to their comparatively higher reconstructed errors. The southwestern Fujian district (China) is chosen as a case study area. A variety of learning rates, iterations, and the size of each hidden layer are constructing and training the deep autoencoder networks on all the geochemical samples. The reconstruction error (or, anomaly score) of each training sample is used to recognize multivariate geochemical anomalies associated with Fe polymetallic mineralization. By comparing the results obtained with a continuous restricted Boltzmann machine, we conclude that the autoencoder network can be trained to recognize multivariate geochemical anomalies. Most of the known skarn-type Fe deposits are located in areas with high reconstruction errors or anomaly scores in the anomaly map, indicating that these anomalies may be related to Fe mineralization. (C) 2015 Elsevier Ltd. All rights reserved.
引用
收藏
页码:75 / 82
页数:8
相关论文
共 67 条
[1]  
AGTERBERG FP, 1990, COMPUT GEOL, V7, P1
[2]   Logratio analysis and compositional distance [J].
Aitchison, J ;
Barceló-Vidal, C ;
Martín-Fernández, JA ;
Pawlowsky-Glahn, V .
MATHEMATICAL GEOLOGY, 2000, 32 (03) :271-275
[3]   Multi-element association analysis of stream sediment geochemistry data for predicting gold deposits in south-central Yunnan Province, China [J].
Ali, Khaled ;
Cheng, Qiuming ;
Li, Wenchang ;
Chen, Yongqing .
GEOCHEMISTRY-EXPLORATION ENVIRONMENT ANALYSIS, 2006, 6 :341-348
[4]  
[Anonymous], 1988, LEARNING REPRESENTAT
[5]  
[Anonymous], 2006, NIPS
[6]  
[Anonymous], 1986, STAT ANAL COMPOSITIO, DOI DOI 10.1007/978-94-009-4109-0
[7]   Learning Deep Architectures for AI [J].
Bengio, Yoshua .
FOUNDATIONS AND TRENDS IN MACHINE LEARNING, 2009, 2 (01) :1-127
[8]  
Bonham-Carter G.F., 1994, Geographic Information Systems for Geoscientists-Modeling with GIS, P416
[9]   Analysis and mapping of geochemical anomalies using logratio-transformed stream sediment data with censored values [J].
Carranza, Emmanuel John M. .
JOURNAL OF GEOCHEMICAL EXPLORATION, 2011, 110 (02) :167-185
[10]   Catchment basin modelling of stream sediment anomalies revisited: incorporation of EDA and fractal analysis [J].
Carranza, Emmanuel John M. .
GEOCHEMISTRY-EXPLORATION ENVIRONMENT ANALYSIS, 2010, 10 (04) :365-381