About this document

Mathematics 14 01662 by Minh Khôi Lê is a document available to read on EtoBox.

This study evaluates the effectiveness of five inferential feature selection methods for predicting item difficulty and discrimination in item response theory (IRT) using machine learning models. Results indicate that distribution-based filters, particularly the Kolmogorov–Smirnov test, significantly enhance predictive accuracy by better capturing the relationship between item features and parameters. The findings emphasize the importance of the distributional properties of item features over the quantity o

Author

Minh Khôi Lê

Language

EN