Ensemble and Boosting-Based Machine Learning Framework for Personality Prediction
Annotatsiya
This study uses ensemble and advanced methods of machine learning to predict traits in individuals. Multiple models, such KNN, Logistic Regression, Random Forest, Gradient Boosting, Cat Boost, XG Boost, and Stacking Classifiers, were employed to improve generalization and durability. Feature engineering and data preparation were included in the pipelines for effective handling. Trained with ideal hyperparameters and a batch size of 32 , the models have always achieved top performance. The greatest performance of the highest performing models was 93% accuracy, with precision of 91%, recall of 92%, and F1-score of 92%. Reduced overfitting was indicated by the lowest training and validation loss. ROC and PR curve tests were used to further indicate stable classification performance. The results show how effective ensemble and boosting techniques are in personality prediction tasks.