Groundwater anomaly detection using machine learning

The purpose of this paper is to present research on modelling and algorithms for groundwater detection. The chosen surrogates are the most correlated features for anomaly detection. The algorithm used in this paper is one class support vector, K- Nearest Neighbors, Gradient Boosting and Decision tree. Colorado River Watch's machine and real-time data are used for putting the model and algorithms through their paces. The Python programme is briefly discussed in order to design the code. Because groundwater contamination is uncommon in practice, we also put our model to the test for anomaly identification using synthetic data from numerical simulations of flow and transport in porous media.