Enhancing Model Interpretability using Local Interpretable Model-Agnostic Explanations (Lime): Insights from Predictive Analysis on Housing Market Dynamics

This paper examines the interpretability of Random Forest Regression models in real estate using Local Interpretable Model-agnostic Explanations (LIME). Analysing LIME explanations for selected instances, we uncover crucial insights into the factors influencing house price predictions. Notably, features such as socioeconomic indicators and environmental metrics significantly impact predicted prices. We highlight the importance of model interpretability for transparency and trust in complex domains like real estate. By employing techniques like LIME, researchers and practitioners can make more informed decisions based on underlying factors. This study contributes to advancing model interpretation, suggesting future research directions for applying LIME in other domains and exploring its effectiveness in diverse predictive models.