![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F26097785%2F47b21230ddc7973b2e4c17b58b6d97f9%2Finbox_26097785_772194a4c4b116e3249952af4cfc228a_Stats%20(4).jpg?generation=1772059847080517&alt=media)
# 📂 Dataset Description

## Bank Customer Churn Dataset

This dataset contains structured customer-level information collected from a retail banking institution. It is designed to simulate a real-world churn prediction scenario where financial behavior, demographic attributes, and account characteristics influence customer retention.

The objective is to predict whether a customer will leave the bank (churn) based on historical records.

---

## 📊 Dataset Files

| File                             | Description                                                        |
| -------------------------------- | ------------------------------------------------------------------ |
| `Bank_Churn.csv`                 | Clean dataset used for modeling                                    |
| `Bank_Churn_Data_Dictionary.csv` | Detailed explanation of each feature                               |
| `Bank_Churn_Messy.xlsx`          | Raw / uncleaned version of the dataset (not required for modeling) |

For this competition, participants should use **Bank_Churn.csv** as the primary modeling dataset.

---

## 📈 Dataset Structure

The dataset consists of customer-level observations including:

* Unique customer identifiers
* Financial metrics
* Demographic attributes
* Account activity indicators
* Target label indicating churn status

### Target Variable

| Column                | Meaning                      |
| --------------------- | ---------------------------- |
| `churn` (or `Exited`) | 1 → Customer left the bank   |
|                       | 0 → Customer remained active |

This is a binary classification problem where the goal is to predict the probability of churn.

---

## 🔎 Feature Categories

The dataset includes a mix of:

### 📌 Numerical Features

* Credit score
* Age
* Account balance
* Estimated salary
* Tenure duration

### 📌 Categorical Features

* Geography
* Gender

### 📌 Binary Indicators

* Has credit card
* Is active member

---

## ⚠ Modeling Considerations

Participants should consider:

* Proper encoding of categorical variables
* Handling potential class imbalance
* Removing non-informative identifiers (e.g., Customer ID, Row Number)
* Preventing data leakage
* Validating models using cross-validation

The dataset is suitable for:

* Logistic Regression
* Random Forest
* Gradient Boosting Models (XGBoost, LightGBM, CatBoost)
* Neural Networks

---

## 🧠 Real-World Context

Customer churn prediction in banking is a high-impact problem. Financial institutions rely on predictive models to:

* Reduce customer attrition
* Improve retention strategies
* Optimize targeted marketing campaigns
* Increase customer lifetime value

This dataset provides a realistic yet structured environment for building and evaluating machine learning models in a business context.

---