Figure 1
The flowchart displays two processes for predictive modelling, labelled a and b. Both start with defining input and target data, followed by data normalisation and random data splitting. For part a, the process involves model training and testing, best parameter selection, and evaluation using metrics such as R-squared, mean squared error, and mean absolute error, leading to the end of prediction. Part b outlines a specific modelling approach that includes training a residual sum model and applying the X G Boost algorithm. It highlights steps such as predicting the residuals and final predictions on the test set. The flowchart features predominantly rectangular boxes, arrows indicating the direction of flow, and a diamond for decision making. The structure remains linear and follows a top-down progression with clearly outlined steps.Architecture of RSM informed (a) ANN modsel and (b) XGBoost model