Evaluation of logistic regression and support vector machine approaches for XRF based particle sorting for a copper ore

被引:16
作者
Xu, Yang [1 ]
Klein, Bern [1 ]
Li, Genzhuang [1 ]
Gopaluni, Bhushan [2 ]
机构
[1] Univ British Columbia, Norman B Keevil Inst Min Engn, 6350 Stores Rd, Vancouver, BC V6T 1Z4, Canada
[2] Univ British Columbia, Chem & Biol Engn, 2360 East Mall, Vancouver, BC V6T 1Z3, Canada
关键词
Sensor -based ore sorting; Receiver operator curve; Logistic regression; Support vector machine; Principal component analysis; LAND-COVER; CLASSIFICATION; PREDICTION; MODELS;
D O I
10.1016/j.mineng.2023.108003
中图分类号
TQ [化学工业];
学科分类号
0817 ;
摘要
The study is aimed at particle sorting at the Copper Mountain Mine using XRF. Possible applications include the rejection of barren material from mill feed, the rejection of pebbles in the SABC circuit or the recovery of valuable material from low grade stockpiles. It is recognized that XRF is a surface measurement that can detect copper but depending on several operational conditions, such as the orientation of the particle or mineral texture, the sensor spot may not see the copper. However, XRF also provides information about the concentrations of a range of elements in minerals that are associated with copper mineralization which can improve sorting. The study described herein is aimed at improving XRF sensor-based sorting by the introduction of logistics regression (LR)-and support vector machine (SVM)-based machine learning approaches. To solve the collinearity and dimensionality issues in the input variables, the authors propose a combined approach of principal component analysis (PCA) and stepwise regression to extract the significant features. The combined PCA and stepwise regression approach is novel and has shown to be very effective for dimensionality reduction of the XRF spectrum data. By applying the ROC and AUC), the LR and SVM models are compared. Results showed that the LR model with the AUC of 0.847 outperforms the SVM with kernel functions with respect to classification accuracy; especially for data sets with a small number of features. The improved classification accuracy should benefit the economic performance of the particle sorting system.
引用
收藏
页数:10
相关论文
共 38 条
[1]   Integrated mining, processing and waste disposal systems for reduced energy and operating costs at Xstrata Nickel's Sudbury Operations [J].
Bamber, A. S. ;
Klein, B. ;
Pakalnis, R. C. ;
Scoble, M. J. .
TRANSACTIONS OF THE INSTITUTIONS OF MINING AND METALLURGY SECTION A-MINING TECHNOLOGY, 2008, 117 (03) :142-153
[2]  
Bamber A.S., 2008, INTEGRATED MINING PR
[3]  
Bender R, 1997, J ROY COLL PHYS LOND, V31, P546
[4]   POINTS OF SIGNIFICANCE Statistics versus machine learning [J].
Bzdok, Danilo ;
Altman, Naomi ;
Krzywinski, Martin .
NATURE METHODS, 2018, 15 (04) :232-233
[5]  
Dalm M, 2011, APPLICABILITY NEAR I
[6]   Logistic regression and artificial neural network classification models: a methodology review [J].
Dreiseitl, S ;
Ohno-Machado, L .
JOURNAL OF BIOMEDICAL INFORMATICS, 2002, 35 (5-6) :352-359
[7]  
Fickling R.S, 2011, 6 SO AFRICAN BASE ME
[8]   Decision tree classification of land cover from remotely sensed data [J].
Friedl, MA ;
Brodley, CE .
REMOTE SENSING OF ENVIRONMENT, 1997, 61 (03) :399-409
[9]   Prediction of grade and recovery in flotation from physicochemical and operational aspects using machine learning models [J].
Gomez-Flores, Allan ;
Heyes, Graeme W. ;
Ilyas, Sadia ;
Kim, Hyunjung .
MINERALS ENGINEERING, 2022, 183
[10]   Variable Importance Assessment in Regression: Linear Regression versus Random Forest [J].
Groemping, Ulrike .
AMERICAN STATISTICIAN, 2009, 63 (04) :308-319