Machine Learning Algorithms and Their Application to Ore Reserve Estimation of Sparse and Imprecise Data (original) (raw)
Related papers
International Conference on Raw Materials and Circular Economy – RawMat2021, 2021
Machine learning algorithms have been used in various steps of mineral resource estimation in the last four decades - from regression algorithms in variogram model fitting to implicit geological modelling using radial basis functions, and grade estimation using artificial neural networks. In most cases reported in scientific literature, machine learning algorithms succeeded to some degree in completing a modelling task - part of a mineral resource estimation study, by outperforming conventional methods either in the time taken to complete the task or the accuracy of the produced results. It is a common claim in most machine learning applications in mineral resource estima-tion, that machine learning algorithms achieve this performance improvement against conven-tional methods, based on less assumptions on the input data distribution and requiring minimum expertise by those who apply them. The speed of current computing systems, personal or cloud based, has allowed for complex models to be built using machine learning algorithms within minutes, leading to a few commercial implementations becoming available to mineral resource estimation practitioners and gaining their acceptance as reliable systems. In the last decade, several mineral resource estimation reports, part of various levels of study from preliminary economic assessments to feasibility studies, were based on the results of machine learning algorithms ap-plication. These reports are commonly released as compliant with one of the internationally ac-ceptable reporting codes, such as JORC or NI 43-101. Therefore, it is important to examine how machine learning algorithms are applied to mineral resource estimation, and how this application complies with the guidelines of international reporting codes for mineral resources, particularly with the requirements for transparency and competence. This paper gives an overview of machine learning algorithms and systems used in mineral resource estimation and discusses possible compliance issues with international reporting codes for mineral resources.
Elsevier, 2019
Predictive modelling of mineral prospectivity using GIS is a valid and progressively more accepted tool for delineating reproducible mineral exploration targets. In this study, machine learning methods, including support vector machine (SVM), artificial neural networks (ANN) and random forest (RF), were employed to conduct GIS-based mineral prospectivity mapping of the Tongling ore district, eastern China. The mineral systems approach was used to translate our understanding of the skarn Cu mineral system into mappable exploration criteria, resulting in 12 predictor maps that represent source, transport, physical trap and chemical deposition processes critical for ore formation. Predictive SVM, ANN and RF models were trained by way of predictor maps, and corroborated using a 10-fold cross-validation. The overall performance of the resulting predictive models was assessed in both training and test datasets using a confusion matrix, set of statistical measurements, receiver operating characteristic curve, and success-rate curve. The assessment results indicate that the three machine learning models presented in this study achieved satisfactory performance levels characterized by high pre-dictive accuracy. In addition, all models exhibited well interpretability that provided consistent ranking information about the relative importance of the evidential features contributing to the final predictions. In comparison, the RF model outperformed the SVM and ANN models, having achieved greater consistency with respect to variations in the model parameters and better predictive accuracy. Importantly, the RF model exhibited the highest predictive efficiency capturing most of the known deposits within the smallest prospective tracts. The above results suggest that the RF model is the most appropriate model for Cu potential mapping in the Tongling ore district, and, therefore, was used to generate a prospectivity map containing very-high, high, moderate, and low potential areas in support of follow-up exploration. The prospective areas delineated in this map occupy 13.97% of the study area and capture 80.95% of the known deposits. The fact that two newly discovered deposits occur within the prospective areas predicted by the prospectivity model indicates that the model is robust and effective regarding exploration target generation.
The Application of Artificial Neural Networks to Ore Reserve Estimation at Choghart Iron Ore Deposit
2017
Geo-statistical methods for reserve estimation are difficult to use when stationary conditions are not satisfied. Artificial Neural Networks (ANNs) provide an alternative to geo-statistical techniques while considerably reducing the processing time required for development and application. In this paper the ANNs was applied to the Choghart iron ore deposit in Yazd province of Iran. Initially, an optimum Multi Layer Perceptron (MLP) was constructed to estimate the Fe grade within orebody using the whole ore data of the deposit. Sensitivity analysis was applied for a number of hidden layers and neurons, different types of activation functions and learning rules. Optimal architectures for iron grade estimation were 3-20-10-1. In order to improve the network performance, the deposit was divided into four homogenous zones. Subsequently, all sensitivity analyses were carried out on each zone. Finally, a different optimum network was trained and Fe was estimated separately for each zone. ...
International Journal of Mining Science and Technology, 2016
In this research, a method called ANNMG is presented to integrate Artificial Neural Networks and Geostatistics for optimum mineral reserve evaluation. The word ANNMG simply means Artificial Neural Network Model integrated with Geostatiscs. In this procedure, the Artificial Neural Network was trained, tested and validated using assay values obtained from exploratory drillholes. Next, the validated model was used to generalize mineral grades at known and unknown sampled locations inside the drilling region respectively. Finally, the reproduced and generalized assay values were combined and fed to geostatistics in order to develop a geological 3D block model. The regression analysis revealed that the predicted sample grades were in close proximity to the actual sample grades. The generalized grades from the ANNMG show that this process could be used to complement exploration activities thereby reducing drilling requirement. It could also be an effective mineral reserve evaluation method that could produce optimum block model for mine design.
Support vector machine: A tool for mapping mineral prospectivity
Computers & Geosciences, 2011
In this contribution, we describe an application of support vector machine (SVM), a supervised learning algorithm, to mineral prospectivity mapping. The free R package e1071 is used to construct a SVM with sigmoid kernel function to map prospectivity for Au deposits in western Meguma Terrain of Nova Scotia (Canada). The SVM classification accuracies of 'deposit' are 100%, and the SVM classification accuracies of the 'non-deposit' are greater than 85%. The SVM classifications of mineral prospectivity have 5-9% lower total errors, 13-14% higher false-positive errors and 25-30% lower false-negative errors compared to those of the WofE prediction. The prospective target areas predicted by both SVM and WofE reflect, nonetheless, controls of Au deposit occurrence in the study area by NE-SW trending anticlines and contact zones between Goldenville and Halifax Formations. The results of the study indicate the usefulness of SVM as a tool for predictive mapping of mineral prospectivity.
Minerals
Spatial prediction of orebody characteristics can often be challenging given the commonly complex geological structure of mineral deposits. For example, a high nugget effect can strongly impact variogram modelling. Geological complexity can be caused by the presence of structural geological discontinuities combined with numerous lithotypes, which may lead to underperformance of grade estimation with traditional kriging. Deep learning algorithms can be a practical alternative in addressing these issues since, in the neural network, calculation of experimental variograms is not necessary and nonlinearity can be captured globally by learning the underlying interrelationships present in the dataset. Five different methods are used to estimate an unsampled 2D dataset. The methods include the machine learning techniques Support Vector Regression (SVR) and Multi-Layer Perceptron (MLP) neural network; the conventional geostatistical methods Simple Kriging (SK) and Nearest Neighbourhood (NN)...
Natural Resources Research, 2010
The aim of this study is to analyze hydrothermal gold-silver mineral deposits potential in the Taebaeksan mineralized district, Korea, using an artificial neural network (ANN) and a geographic information system (GIS) environment. A spatial database considering 46 Au and Ag deposits, geophysical, geological, and geochemical data was constructed for the study area using the GIS. The geospatial factors were used with the ANN to analyze mineral potential. The Au and Ag mineral deposits were randomly divided into a training set (70%) to analyze mineral potential using ANN and a test set (30%) to validate predicted potential map. Four different training datasets determined from likelihood ratio and weight of evidence models were applied to analyze and validate the effect of training. Then, the mineral potential index (MPI) was calculated using the trained back-propagation weights, and mineral potential maps (MPMs) were constructed from GIS data for the four training cases. The MPMs were then validated by comparison with the test mineral occurrences. The validation results gave respective accuracies of 73. 06, 73.52, 70.11, and 73.10% for the training cases. The comparison results of some training cases showed less sensitive to training data from likelihood ratio than weight of evidence. Overall, the training cases selected from 10% area with low and high index value of MPM L and MPM W gave higher accuracy (73.52 and 73.10%) for MPMs than those (73.06 and 70.11%, respectively) from known deposits and 10% area with low index value of MPI L and MPI W .
2005 International Conference on Machine Learning and Cybernetics, 2005
In the mining industry, effective use of geographic information systems (GIS) to identify new geographic locations that are favorable for mineral exploration is very important. However, definitive prediction of such location is not an easy task. In this paper, four different neural networks, namely, the Polynomial Neural Network (PNN), General Regression Neural Network (GRNN), Probabilistic Neural Network (PrNN) and Back Propagation Neural Network (BPNN) have been used to classify data corresponding to cells in a map grid into deposit cells and barren cells. These approaches were tested on the GIS mineral exploration data from the Kalgoorlie region of Western Australia. The performance of individual neural networks is compared based on simulation results. The results demonstrate various degrees of success for the networks and suggestions on how to integrate the results are discussed.