| Περίληψη: |
When conventional supervised machine learning (ML), including decision trees, support vector machines (SVMs), and k‐nearest neighbors (KNN), is applied to geological problems involving complex data sets, it is necessary to select a subset of raw or pre‐processed data types that will be used as input to the ML model. We revised four ML case studies involving 2D and 3D structural and lithological inference from geophysical survey data (magnetic, gravity, and radiometric measurements). For each study, we identified the most relevant inputs via a package of input selection approaches, comprising descriptive statistics, principal component analysis (PCA), correlation coefficient analysis, significance testing, and algorithmic input selection techniques, including Pearson, Spearman, Kendall, Minimum Redundancy Maximum Relevance (MRMR), Relief, permutation importance, Local Interpretable Model‐agnostic Explanations (LIME), and Shapley values. As anticipated, strategic input selection reduced collinearity among the raw input data sets and their standard derivatives, and consistently enhanced model performance across all case studies, improving accuracy, precision, and recall while reducing overfitting. However, different input selection methods proved optimal in different case studies, with no single approach consistently outperformed others across all geological contexts. This demonstrates the importance of using multiple complementary input selection methods when developing ML applications for automated geological mapping. When applying ML models to new localities with different geological contexts, data‐driven input selection and feature engineering should be revisited to ensure model performance, rather than assuming direct transferability of input configurations from the original study area. Plain Language Summary: We explore how to choose the best data inputs when using machine learning to solve geological problems. We looked at four different geological scenarios, trying to predict geological structures and rock types using various geophysical measurements. We tested different ways of picking the most useful data inputs and found that carefully selecting inputs improved the accuracy of their predictions. We found that processed versions of the original data often worked better than raw measurements. We also found that no single method of choosing inputs worked best for all scenarios. Therefore, we suggest that when applying machine learning to new geological problems, it's important to try multiple methods for selecting data inputs to get the best results. Key Points: Multiple algorithmic input selection techniques were applied to four geological machine learning (ML) case studies using geophysical survey dataCareful input selection improves model performance. Processed and filtered data often outperform raw data as inputsNo single input selection method was best across all cases, suggesting the need for a multi‐method approach in geological ML [ABSTRACT FROM AUTHOR] |