Bank Client Retention Forecasting Using ANN

[sanskaryo]

Deep Learning Simplified Repository (Proposing new issue)

🔴 Project Title : Customer Retention Risk Prediction

🔴 Aim : Predict customer churn risk in the banking sector to help retain high-risk customers by identifying patterns through factors like credit score, balance, and activity levels.

🔴 Dataset : Real-world banking dataset containing features such as customer age, credit score, balance, and tenure.

🔴 Approach : Use at least 3-4 algorithms including an Artificial Neural Network (ANN) for model implementation. Compare the models based on accuracy scores, and perform exploratory data analysis (EDA) to understand key trends and data distributions before building the models.

---

✅ To be Mentioned while taking the issue :

• Full name: Sanskar khandelwal
• GitHub Profile Link: https://github.com/sanskaryo
• Email ID: [email protected]
• Participant ID (if applicable):
• Approach for this Project: pproach: Preprocess the data, perform EDA, build and compare models ANN, and analyze feature importance. Integrate with Streamlit for predictions, and optionally provide Streamlit deployment links for easy access.
• What is your participant role? - gssoc-ext and hacktoberfest

---

Thankyou

[github-actions]

Thank you for creating this issue! We'll look into it as soon as possible. Your contributions are highly appreciated! 😊

[abhisheks008]

Assigned @sanskaryo

Make sure you implement at least 4 different deep learning models for this problem statement otherwise this will not be considered.

[sanskaryo]

Ok Sir , Thanks for assigning

[mistbik]

Hey, could I be assigned this?

Goals: to preprocess the data, perform EDA to identify trends/patterns in the data, implement several algorithms (including ANN) for my model, (compare accuracy of said algorithms) and predict customer activity through factors such as credit, balance, activity.

[abhisheks008]

Hi @mistbik sorry for the huge delay.

You can start working in this issue. Please try to implement at least 3-4 deep learning models for this dataset.