📊 Customer Churn Prediction & Business Analytics Platform

An end-to-end Data Analytics and Machine Learning project that analyzes customer behavior, identifies churn patterns, and predicts customers who are likely to leave a telecom service.

The project combines Python, PostgreSQL, Machine Learning, and Tableau to perform data preprocessing, business analysis, and churn prediction.

Customer churn is one of the biggest challenges for subscription-based businesses. This project analyzes historical customer data to discover the key factors influencing churn and builds machine learning models to predict customer attrition.

• Python
• PostgreSQL
• Scikit-learn
• XGBoost
• Pandas
• NumPy
• Matplotlib
• Seaborn
• Tableau
• Jupyter Notebook

Customer_Churn_Analytics/

├── Dataset/
│   ├── raw/
│   └── processed/
│
├── Python/
│   ├── 01_Data_Understanding.ipynb
│   ├── 02_Data_Cleaning.ipynb
│   └── 03_EDA.ipynb
│
├── SQL/
│   ├── 01_Create_Database.sql
│   ├── 02_Create_Tables.sql
│   ├── 03_Data_Import.sql
│   ├── 04_Data_Validation.sql
│   ├── 05_Business_Analysis.sql
│   └── 06_Dashboard_Views.sql
│
├── ML_Model/
│   ├── 01_Feature_Engineering.ipynb
│   └── 02_Model_Training.ipynb
│   ├── 03_Model_Evaluation.ipynb
├── Tableau/
│
├── Images/
│
├── requirements.txt
└── README.md

Raw Dataset
      │
      ▼
Data Understanding
      │
      ▼
Data Cleaning
      │
      ▼
Exploratory Data Analysis
      │
      ▼
SQL Business Analysis
      │
      ▼
Feature Engineering
      │
      ▼
Machine Learning Model Training
      │
      ▼
Business Dashboard (Tableau)

• Data Understanding
• Data Cleaning
• Exploratory Data Analysis (EDA)
• SQL Business Analysis
• Feature Engineering
• Model Training

• Model Evaluation
• Customer Risk Scoring
• Business Insights Report
• Final Project Documentation

• Logistic Regression
• Random Forest Classifier
• XGBoost Classifier

Source: IBM Telco Customer Churn Dataset

The dataset contains customer demographics, subscription details, billing information, and churn status.

Build an end-to-end analytics solution capable of:

• Understanding customer behavior
• Identifying churn patterns
• Supporting business decision-making
• Predicting customer churn using Machine Learning

Devi Sri Parvathi Junjuri

B.Sc. Computer Science Student

⭐ If you found this project useful, consider giving this repository a star.