Telecom_Customer_Churn_Prediction

Overview

This project focuses on predicting customer churn for a telecommunications company. Customer churn, also known as customer attrition, refers to customers who discontinue their services with a company. Understanding and predicting churn is crucial for telecom businesses as it enables them to take proactive measures to retain valuable customers. In this particular project I tried to recommend a discount of up to 20%, depending to how likely a customer will leave. check model prediction here: (https://telecomcustomerchurnprediction-v1.streamlit.app/)

Project Goals

1. Data Exploration and Analysis: Thoroughly investigate the customer dataset to identify patterns, trends, and potential factors influencing churn.
2. Model Building: Develop predictive models to classify customers as likely or unlikely to churn.
3. Model Evaluation: Assess model performance using appropriate metrics and validation techniques.
4. Insights and Recommendations: Provide actionable insights and recommendations based on model results to reduce churn rates.

Dataset

The project uses the "Telco Customer Churn" dataset. This dataset includes information on customer demographics, service usage, contract details, and churn status (whether the customer churned or not).

Key Features

• Customer Demographics: Gender, age, marital status, dependents.
• Service Usage: Phone service, multiple lines, internet service, online security, online backup, device protection, tech support, streaming TV, streaming movies.
• Contract Details: Contract type (month-to-month, one year, two year), payment method, paperless billing.
• Charges: Monthly charges, total charges.
• Churn Label: Whether the customer churned or not.

Methodology

1. Exploratory Data Analysis (EDA):
   • Visualize distributions and relationships between features.
   • Identify potential correlations with numerical features.
2. Data Preprocessing:
   • Handle missing values.
   • Encoding categorical variables.
   • Scaling numerical features.
3. Model Selection and Training:
   • Experiment with various classification models (e.g., Logistic regression, Random forest, eXtreme Gradient Boosting, Support Vector Machine, Neural network).
   • Tune hyperparameters for optimal performance.
4. Model Evaluation:
   • Use metrics like accuracy, precision, recall, F1-score, and AUC-ROC.
   • Employ cross-validation to assess model robustness.

Additional Notes

• This project is intended for educational and exploratory purposes.
• The models developed here may not be directly applicable to some scenarios without further customization and validation.

Visuals of Model at work