Explainable Neural Network Regression Model with SHAP

Radial Basis Function Neural Network training include 5-fold cross-validation and SHAP analysis for explainable model

This MATLAB script implements an explainable neural network regression model using a Radial Basis Function Neural Network (RBFNN) to predict water flux in forward osmosis processes. The model utilizes operational parameters such as membrane area, feed and draw solution flow rates, and concentrations as input features for training. To enhance interpretability, SHapley Additive exPlanations (SHAP) are applied, allowing users to gain insights into the contribution of each parameter to the model's predictions. This tool provides a powerful solution for researchers and engineers looking to develop accurate and transparent regression models while leveraging the flexibility of RBFNNs for optimizing forward osmosis system performance.