🚀 BankPulse Pro — Loan Acceptance Intelligence

An advanced Machine Learning + Luxury Analytics Dashboard for predicting customer acceptance of personal loan offers using the UCI Bank Marketing Dataset.

🧠 Project Overview

This project builds an end-to-end classification system that:

Predicts whether a customer will accept a loan offer Provides real-time prediction via Streamlit dashboard Visualizes customer segments, campaign performance, and insights Compares Logistic Regression vs Decision Tree models

👉 Built as a portfolio-grade SaaS-style dashboard with enterprise-level UI.

🎯 Objective

Predict which customers are more likely to accept a personal loan offer based on:

Demographics (age, job, marital status) Financial data (balance, loans) Campaign interaction history 📂 Dataset Source: UCI Machine Learning Repository Dataset: Bank Marketing Dataset File: bank-full.csv ⚙️ Tech Stack Python 🐍 Pandas / NumPy Scikit-learn Streamlit (Frontend Dashboard) Plotly (Interactive Visualizations) Joblib (Model Serialization) 🧪 Models Used 🔹 Logistic Regression Balanced class weights Scaled numerical features Strong baseline performance 🌳 Decision Tree Classifier Controlled depth to prevent overfitting Provides feature importance insights 📊 Evaluation Metrics Accuracy Precision Recall F1-Score ROC-AUC 🏗️ Project Structure 📁 project/ │ ├── train.py # Model training & evaluation ├── app.py # Streamlit dashboard ├── data/ │ └── bank-full.csv # Dataset │ ├── lr_model.pkl # Logistic Regression model ├── dt_model.pkl # Decision Tree model ├── scaler.pkl # StandardScaler ├── columns.pkl # Feature columns ├── numeric_features.pkl # Numeric feature list 🚀 How to Run 1️⃣ Install dependencies pip install pandas numpy scikit-learn streamlit plotly joblib 2️⃣ Train the models python train.py 3️⃣ Run the dashboard streamlit run app.py 🖥️ Dashboard Features

This project includes a premium SaaS-style dashboard:

📊 Executive Dashboard KPI cards (Conversion Rate, Revenue Potential) Customer segmentation charts Campaign performance trends 🔮 Predictor Module Real-time customer input Loan acceptance probability Risk analysis insights 📈 Analytics Feature importance visualization Age distribution & acceptance trends Customer segmentation matrix ⚙️ Model Insights Model comparison (LR vs Decision Tree) Confusion matrix visualization Business intelligence insights 💡 Key Business Insights 🎯 Students & retirees show highest acceptance rates 💰 Account balance is strongest predictor 📞 Too many contacts reduce conversion 📅 Campaign timing affects success 🔄 Previous success increases acceptance probability 🎨 UI Highlights Luxury fintech-inspired design Dark/Light theme switching Glassmorphism & gradient UI Interactive charts (Plotly) Enterprise dashboard layout 📌 Example Output Acceptance Probability: 72.4% Decision: ACCEPTED / DECLINED Risk Factors: Highlighted dynamically 📈 Skills Demonstrated Machine Learning (Classification) Feature Engineering Model Evaluation Data Visualization Streamlit App Development Business Insight Extraction 🔥 Future Improvements Add XGBoost / LightGBM Hyperparameter tuning (GridSearchCV) Model deployment (FastAPI + Docker) User authentication (SaaS version) Real-time API integration 👨‍💻 Author

Mudassir Hassan Data Science & AI Enthusiast

⭐ If You Like This Project

Give it a ⭐ on GitHub and share feedback!