https://github.com/sumdiboii/loan-prediction-decision-trees

A Decision Tree Classifier was implemented to predict personal loan acceptance using a dataset of 5,000 customers. Key features included income, education, mortgage, and credit card usage. The model achieved 97% accuracy, with 92% precision and 76% recall for positive loan predictions, validated using a classification report and confusion matrix.
https://github.com/sumdiboii/loan-prediction-decision-trees

classification data-visualisation decision-trees loan-prediction machine-learning python scikit-learn supervised-learning

Last synced: about 2 months ago
JSON representation

A Decision Tree Classifier was implemented to predict personal loan acceptance using a dataset of 5,000 customers. Key features included income, education, mortgage, and credit card usage. The model achieved 97% accuracy, with 92% precision and 76% recall for positive loan predictions, validated using a classification report and confusion matrix.

• Host: GitHub
• URL: https://github.com/sumdiboii/loan-prediction-decision-trees
• Owner: Sumdiboii
• Created: 2025-06-21T14:45:57.000Z (about 1 year ago)
• Default Branch: main
• Last Pushed: 2025-06-21T15:17:54.000Z (about 1 year ago)
• Last Synced: 2025-06-21T15:35:29.302Z (about 1 year ago)
• Topics: classification, data-visualisation, decision-trees, loan-prediction, machine-learning, python, scikit-learn, supervised-learning
• Language: Jupyter Notebook
• Homepage: https://colab.research.google.com/drive/1tUx7J-Nv1iLBHVT84zBBrdLQOzSLgNS0?usp=sharing
• Size: 0 Bytes
• Stars: 0
• Watchers: 0
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

          



  





  






💼 Loan Prediction Using Decision Trees 💼

Scikit-learn powered classification model for banking analytics.







  

    

  






---




## 🚀 Project Overview

This project utilizes a Decision Tree Classifier to predict whether a customer will accept a personal loan offer, based on a dataset of 5,000 banking records. Key features include customer income, education level, credit card spending, mortgage value, and account behavior. The model was trained using scikit-learn and evaluated with a classification report and confusion matrix, achieving an **accuracy of 97%**, with **92% precision** and **76% recall** for predicting positive loan responses.




---




## 🛠️ Tech Stack and Tools






  

    

      Technology

      Purpose

    

  

  

    

      Python

      Primary programming language

    

    

      Pandas

      Data loading, cleaning, and manipulation

    

    

      Matplotlib & Seaborn

      Data visualization and analysis

    

    

      Scikit-learn

      Model training, testing, and evaluation

    

    

      Jupyter Notebook / Colab

      Interactive development and visualization

    

  






---




## 🔍 Core Highlights

- 📊 Cleaned and preprocessed 5,000 rows of real-world banking data  

- 🌲 Trained a Decision Tree Classifier with `max_leaf_nodes=3`  

- 🧠 Achieved **97% overall accuracy**  

- 🎯 Precision: **92%** for identifying loan-accepting customers  

- 📉 Visualized loan acceptance trends across education, family size, and financial behavior  




---




## 📚 Key Learning Outcomes

- Practical understanding of classification algorithms using Decision Trees  

- Insights into data-driven loan decision modeling  

- Experience in evaluating model performance with precision, recall, and confusion matrix  

- Visualization of model errors using Seaborn  




---




## 👨‍💻 About the Creator

**Sumdiboii** – Machine Learning Enthusiast & Software Developer  

*LinkedIn – [Sumedh Pimplikar](https://www.linkedin.com/in/sumedh-pimplikar)*




> **From raw banking data to sharp predictions — this project showcases the practical power of decision trees in solving real-world business problems.**