Enhanced Explanations in Recommendation Systems

Recommendation Systems (RSs) play a crucial role in assisting users in making decisions and finding their desired items in various domains, such as movies, music, and hotels. However, their complex algorithms often raise concerns about transparency, fairness, and user trust. To address these challenges, we tried to propose a theoretical approach that combines SHAP (SHapley Additive exPlanations) and LIME (Local Interpretable Model-Agnostic Explanations) techniques to enhance the transparency and interpretability of RSs. We present a methodology for applying this approach in the context of movie recommendation, where SHAP values quantify global feature importance, and LIME explanations provide localized insights. This work can contribute to the advancement of transparent and user-centric RSs, with implications for a wide range of applications.