Modeling Semiarid River–Aquifer Systems with Bayesian Networks and Artificial Neural Networks

In semiarid areas, precipitations usually appear in the form of big and brief floods, which affect aquifer through water infiltration, causing groundwater temperature changes. These changes may have an impact on physical, chemical biological processes and, thus, modeling variations associated with stormy precipitation episodes is essential, especially since this kind becoming increasingly frequent regions. paper, we compare predictive performance two popular tools statistics machine learning, namely Bayesian networks (BNs) artificial neural (ANNs), variation events. More specifically, trained a total 2145 ANNs different node configurations, from one to five layers. On other hand, three BNs using structure learning algorithms. We conclude that, while both are equivalent terms accuracy for predicting drops, computational cost estimation significantly lower, resulting BN models more versatile allow detailed analysis.