Machine Learning-Aided Three-Dimensional Geological Modeling with Uncertainty Quantification

Three-dimensional (3D) geological modeling plays a crucial role in understanding and predicting subsurface features. This paper presents a machine learning-aided approach for 3D geological modelingto predict geological units at unsampled locations. Four popular machine learning algorithms, support vector machine (SVM),k-nearest neighbours (kNN), gradient boosting decision tree (GBDT) and random forest (RF), are compared for their performance using confusion matrix andF1 score. The concept of information entropy is introduced to quantify uncertainty in the geological models. The proposed approach is applied to a site in Hong Kong. The distributions of underground geological layers are created from the developed model. Machine learning has shown promising applicability in 3Dgeological modeling.

Keywords: 3D geological modelling, Machine learning, Uncertainty.