Search

The different methods for delineating favorable areas for mineral exploration utilize exploration criteria regarding targeted mineral deposits. The criteria are elicited according to conceptual model parameters of the targeted mineral deposits. The selection of indicator criteria, the evaluation of their comparative importance, and their integration are critical in mineral prospectivity modelling. In data-driven methods, indicator features are weighted using functions whereby the importance of certain indicator criteria may be ignored. In this paper, a datadriven method is described for recognizing and converting exploration criteria into quantitative coefficients representing favorability for the presence of the targeted mineral deposits. In this approach, all indicator features of the targeted mineral deposits are recognized and incorporated in the modelling procedure. The method is demonstrated for outlining favorable areas for a Mississippi valley-type fluorite deposit in an area, north of Iran. The method is developed by studying and modelling the geological characteristics of known mineral occurrences. The degree of prediction ability of each exploration criterion is quantified as a recognition coefficient, which can be used as a weight attributed to the criterion in mineral exploration targeting to outline favorable areas. [ABSTRACT FROM AUTHOR]

[Display omitted] • Majority voting ensemble technique is proposed for mineral prospectivity analysis. • Confidence index is proposed to mitigate uncertainty associated with prospectivity models. • The methods facilitate robust extraction of exploration targets. • The proposed approaches can be included in exploration information system (EIS). • The framework deals with problem solving in mineral exploration targeting. Various mineral prospectivity modelling (MPM) approaches are available for targeting mineral deposits, each method capable of predicting areas of high prospectivity. Given the diversity of MPM approaches, the modelled areas of high prospectivity can differ across different MPMs. However, rather than a negative, different MPM outputs can benefit mineral exploration targeting because each method has its advantages. Rather, the problem lies in the lack of consensus over how to best select and delimit mineral exploration targets from different MPM results. Here we aim to address the challenges outlined above whilst quantifying and mitigating the effects of inherent uncertainties. We first generate eleven different prospectivity models utilising deep learning, machine learning, fuzzy logic, and geometric average integration methods. Then, we adopt a majority voting ensemble technique to incorporate and combine the predictions of each prospectivity model. Next, we propose a confidence index designed to mitigate uncertainty associated with our multi-technique approach to MPM. The confidence index quantifies variation in prospectivity values for each cell of the MPM target area. The conjunction of a confidence index and majority voting model facilitates consistent and robust algorithm-driven extraction of exploration targets based on an ensemble of prospectivity models. [ABSTRACT FROM AUTHOR]

In mineral exploration programs, reducing uncertainty and increasing exploration success have always been challenging issues. To modulate the above-mentioned uncertainty and increase exploration accomplishment, integration, and prospectivity analysis techniques are used for mineral exploration targeting. This paper aims to model the mineral potential of porphyry copper deposits in the Jiroft region, Kerman province. To achieve this goal and overcome the aforementioned issues resulting from the operation of complex ore-forming geological processes, continuous weighting methods through logistic functions were used while training points and analyst's opinions were not contributed to the weighting procedure. Then, to generate exploration targets, the weighted layers were combined with three different integration methods namely, artificial neural network, geometric average, and fuzzy gamma operators. The comparison of the model obtained from the application of an artificial neural network with those obtained by the geometric average and the fuzzy gamma operators using prediction rate-area plots indicated that all the models have good overall performance and acceptable prediction rate. However, the performance of the artificial neural network model is slightly less than that of the other two models. Thus, the targets generated using the geometric average and fuzzy gamma operators are more reliable for planning further exploration programs. [ABSTRACT FROM AUTHOR]

Geochemical exploration data play a vital role in mineral prospectivity modelling (MPM) for discovering unknown mineral deposits. In this study, the improved spatially weighted singularity mapping (SWSM) method is used to improve the practice of identifying geochemical anomalies related to copper mineralization in the Sarduiyeh district, Iran. Then, the random forest algorithm (RF) and geometric average function (GA) are used to integrate the resulting geochemical predictor map with other predictor maps. As demonstrated by the high area under the curve (AUC) values, this approach can effectively delineate prospective areas with RF and GA. However, compared to the GA approach (AUC=0.78), the RF technique (AUC = 0.98) offers superior prediction capabilities due to its enhanced ability to capture spatial correlations between predictive maps and known mineral deposits. The proposed procedure, a hybrid of the improved SWSM and RF has outstanding predictive capabilities for identifying prospective areas. A case in point is the new, highly prospective areas identified in this study, which present priority targets for future exploration in the Sarduiyeh district. [ABSTRACT FROM AUTHOR]

This paper uses support vector machine (SVM), back propagation neural network (BPNN), and Multivariate Regression Analysis (MLA) methods to predict the gold in the Dalli deposit situated in the central province of Iran. After analyzing the data, the dataset was prepared. Subsequently, through comprehensive statistical analyses, Au was chosen as the output element for modelling, while Cu, Al, Ca, Fe, Ti, and Zn were considered input parameters. Then, the dataset was divided into two groups: training and testing datasets. For this purpose, 70% of the datasets were randomly entered into the data process, while the remaining data were assigned for the testing stage. The correlation coefficients for SVM, BPNN, and MLA were 94%, 75%, and 68%, respectively. A comparison of these coefficients revealed that all used methods successfully predicted the actual grade of Au. However, the SVM was more reliable and accurate than other methods. Considering the sensitivity of the gold data and the small number in the exploratory database, the results of this research are used to prepare the main layer in the mineral prospectivity mapping (MPM) of gold in 2 and 3D. [ABSTRACT FROM AUTHOR]