Fig. 1
From: Development and validation of machine learning models for predicting blastocyst yield in IVF cycles
Performance comparison of machine learning models using recursive feature elimination (RFE). The figure illustrates the impact of RFE on model performance across four machine learning algorithms: Light Gradient Boosting Machine (LightGBM), Linear Regression (LR), Support Vector Machine (SVM), and Extreme Gradient Boosting (XGBoost). Features are systematically eliminated from 21 down to 2. The top panel presents the test R2 (coefficient of determination), where higher values indicate better model fit, while the bottom panel displays the test Mean Absolute Error (MAE), where lower values represent better prediction accuracy.
