Fuzzy Gaussian Copula SDG

Fuzzy Gaussian Copula synthetic data generation

The provided MATLAB code integrates fuzzy logic into the processing of the Iris dataset by generating synthetic data through Gaussian copula and transforming both original and synthetic features into fuzzy membership scores using a Gaussian membership function. These scores are then used to train decision tree models separately on the original, synthetic, and combined datasets. Train-test splits ensure the models are evaluated on unseen data to test their generalizability. The code handles the entire workflow from data manipulation to final evaluation, presenting accuracies for models trained on different data sets. This approach is intended to leverage fuzzy logic for better management of uncertainty and noise in the data, potentially enhancing the robustness and performance of the classification models.