[PDF]

Explainable Loan Approval System


Hamed Al-Baklani

14/05/2025

Supervised by Alexia Zoumpoulaki; Moderated by Crispin Cooper

Using open datasets, develop an (online) application that allows you to load data and train a model that predicts loan approvals based on various input parameters, and more importantly, provides clear and understandable reasons for its decisions. Data Collection: Gather datasets with features like credit score, income, employment history, loan amount, etc. Model Development: Use a ML models like a Decision Tree or Random Forest for its inherent explainability. Explainability: Integrate SHAP (SHapley Additive exPlanations) or LIME (Local Interpretable Model-agnostic Explanations) to decipher the model decisions. UI Development: Build a simple user interface where users can input their data and receive both a loan prediction and an explanation for the decision.


Initial Plan (03/02/2025) [Zip Archive]

Final Report (14/05/2025) [Zip Archive]

Publication Form