4. Experiment 4.1. Evaluation Index The 10-fold cross-validation method was used, in which the dataset was divided into 10 folds and evaluated. A total of 10% of the total data was held out and used as test data, while the remaining 90% of the data were used for training and validation. Since the training and validation data change from fold to fold during cross-validation, the performance of the model is not dependent on the specific dataset, but rather generalized.
