Exploring the Relationship Between Feature Attribution Methods and Model Performance
This work addresses the gap in understanding explainability factors in educational machine learning models, but it is incremental as it applies existing methods to analyze correlations.
The study investigated the relationship between feature attribution methods and model performance in predicting student success, finding a very strong correlation between model performance and the agreement among explanation methods.
Machine learning and deep learning models are pivotal in educational contexts, particularly in predicting student success. Despite their widespread application, a significant gap persists in comprehending the factors influencing these models' predictions, especially in explainability within education. This work addresses this gap by employing nine distinct explanation methods and conducting a comprehensive analysis to explore the correlation between the agreement among these methods in generating explanations and the predictive model's performance. Applying Spearman's correlation, our findings reveal a very strong correlation between the model's performance and the agreement level observed among the explanation methods.