『Abstract
　Data-driven prospectivity mapping can be undermined by dissimilarity in multivariate spatial data signatures of deposit-type location. Most cases of data-driven prospectivity mapping, however, make use of training sets of randomly selected deposit-type locations with the implicit assumption that they are coherent (i.e., with similar multivariate spatial data signatures). This study shows that the quality of data-driven prospectivity mapping can be improved by using a training set of coherent deposit-type location. Analysis and selection of coherent deposit-type locations was performed via logistic regression, by using multiple sets of deposit occurrence favourability scores of univariate geoscience spatial data as independent variables and binary deposit occurrence scores as dependent variable. The set of coherent deposit-type locations and three sets of randomly selected deposit-type locations were each used in data-driven prospectivity mapping via application of evidential belief functions. The prospectivity map based on the training set of coherent deposit-type locations resulted in lower uncertainty, better goodness-of-fit to the training set, and better predictive capacity against a cross-validation set of economic deposits of the type sought. This study shows that explicit selection of training set of coherent deposit-type locations should be applied in data-driven prospectivity mapping.

Keywords: Cumulative frequency distributions; Logistic regression; Evidential belief functions; GIS; Alkalic porphyry Cu-Au; Structural controls; Geophysics; Geochemistry; British Columbia (Canada)』

1. Introduction
2. The test area and geoscience spatial data sets
3. Mineral occurrence favourability scores of univariate geoscience spatial data
　3.1. Deposits excluded from the analysis
　3.2. Univariate derivative geoscience spatial data used in the analysis
　3.3. Analysis of spatial association
　3.4. Results and discussion
4. Analysis and selection of coherent deposit locations
　4.1. Concept
　4.2. Logistic regression analysis
　4.3. Results and discussion
5. Data-driven prospectivity mapping
　5.1. Evidential belief functions
　5.2. Training deposit locations
　5.3. Thematic evidential data layers
　5.4. Data-driven estimation and integration of EBFs
　5.5. Estimated EBFs of evidential data layers and integrated EBFs
　5.6. Performance of prospectivity maps
6. Discussion
7. Conclusions
References