A positive and unlabeled learning algorithm for mineral prospectivity mapping

被引:43
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
Positive and unlabeled learning; One-class classification; Machine learning; Mineral prospectivity mapping; ONE-CLASS CLASSIFICATION; SUPPORT VECTOR MACHINE; MAKENG FE DEPOSIT; RANDOM FORESTS; LOGISTIC-REGRESSION; TEXT CLASSIFICATION; SAMPLING STRATEGIES; FUJIAN PROVINCE; NEURAL-NETWORK; PREDICTION;
D O I
10.1016/j.cageo.2020.104667
中图分类号
TP39 [计算机的应用];
学科分类号
081203 ; 0835 ;
摘要
Application of supervised machine learning algorithms for mineral prospectivity mapping (MPM) requires positive and negative training samples. Typically, known mineral deposits are considered as positive training samples. However, the selection of negative training samples in the process of MPM is challenging. The one-class classification methods require positive and unlabeled samples or only positive samples; while without requiring negative training samples. In this study, the positive and unlabeled learning (PUL) algorithm was employed to produce a potential map for Fe polymetallic mineralization in southwestern Fujian province, China. This study first examined the sensitivity of the PUL algorithm to different training sets of labeled and unlabeled locations. The predictive results on 10 random unlabeled datasets confirm that PUL modeling with different training sets is reproducible and stable. In addition, the trained model provides a strong spatial correlation between the predictive variables and the locations of known mineral deposits. Finally, the performance of PUL1algorithm is compared to one-class support vector machine (OCSVM), artificial neural networks (ANN), and logistic regression (LR). Comparative results indicate that the PUL model can achieve a better performance in terms of both fitting-rate, prediction-rate, and AUC value compared with OCSVM, ANN and LR. The labelling efforts can be significantly reduced because the PUL algorithm requires only a small number of positive samples and utilizes unlabeled data in training.
引用
收藏
页数:9
相关论文
共 94 条
[1]  
[王森 Wang Sen], 2015, [地球化学, Geochimica], V44, P450
[2]  
Agterberg F. P., 1999, P 28 INT S APPL COMP, V483, P490
[3]  
Agterberg F.P., 2005, Nat. Resour. Res., V14, P1, DOI DOI 10.1007/S11053-005-4674-0
[4]   Applying a sophisticated approach to predict CO2 solubility in brines: application to CO2 sequestration [J].
Ahmadi, Mohammad Ali ;
Ahmadi, Alireza .
INTERNATIONAL JOURNAL OF LOW-CARBON TECHNOLOGIES, 2016, 11 (03) :325-332
[5]   Robust intelligent tool for estimating dew point pressure in retrograded condensate gas reservoirs: Application of particle swarm optimization [J].
Ahmadi, Mohammad Ali ;
Ebadi, Mohammad ;
Yazdanpanah, Arash .
JOURNAL OF PETROLEUM SCIENCE AND ENGINEERING, 2014, 123 :7-19
[6]   Evolving predictive model to determine condensate-to-gas ratio in retrograded condensate gas reservoirs [J].
Ahmadi, Mohammad Ali ;
Ebadi, Mohammad ;
Marghmaleki, Payam Soleimani ;
Fouladi, Mohammad Mahboubi .
FUEL, 2014, 124 :241-257
[7]   Evolving artificial neural network and imperialist competitive algorithm for prediction oil flow rate of the reservoir [J].
Ahmadi, Mohammad Ali ;
Ebadi, Mohammad ;
Shokrollahi, Amin ;
Majidi, Seyed Mohammad Javad .
APPLIED SOFT COMPUTING, 2013, 13 (02) :1085-1098
[8]   New approach for prediction of asphaltene precipitation due to natural depletion by using evolutionary algorithm concept [J].
Ahmadi, Mohammad Ali ;
Shadizadeh, Seyed Reza .
FUEL, 2012, 102 :716-723
[9]   Connectionist model predicts the porosity and permeability of petroleum reservoirs by means of petro-physical logs: Application of artificial intelligence [J].
Ahmadi, Mohammad-Ali ;
Ahmadi, Mohammad Reza ;
Hosseini, Seyed Moein ;
Ebadi, Mohammad .
JOURNAL OF PETROLEUM SCIENCE AND ENGINEERING, 2014, 123 :183-200
[10]   A rigorous model to predict the amount of Dissolved Calcium Carbonate Concentration throughout oil field brines: Side effect of pressure and temperature [J].
Ahmadi, Mohammad-Ali ;
Bahadori, Alireza ;
Shadizadeh, Seyed Reza .
FUEL, 2015, 139 :154-159