Algorithm 1. Feature selection using Lasso Regression.

The input is as follows:      $X$: Design matrix containing predictor variables (features).      $y$: Vector of observed target values.      $\lambda$: Regularization parameter for LASSO regression.      $k$: Number of folds for cross-validation. 2. Standardize the data:       i. Center each feature by subtracting its mean.      ii. Scale each feature by dividing by its standard deviation. 3.Initialize an empty list to store selected features. 4. Perform k-fold cross-validation:       i. Split the data into k equal-sized folds.      ii. For each fold: a Use the remaining (k − 1) folds as the training set and the current fold as the validation set. b Fit a LASSO regression model on the training data. c Apply an appropriate metric to the validation set to assess the efficacy of the model. d Record the coefficients of the LASSO model. 5.Find the common measure of efficiency for all folds for each value of $\lambda$. 6.Select the optimal value of $\lambda$ that minimises the performance. 7.Fit a LASSO regression model on the entire dataset using the selected $\lambda$. 8.Extract the coefficients of the LASSO model. 9.Identify the features with non-zero coefficients and add them to the list of selected features. 10. The output is as follows:       i. List of selected features.