🤖ML Model

Predicting Customer Churn Model

This project involves building a robust machine learning model to predict customer churn for a telecommunications company. The workflow includes comprehensive data cleaning, exploratory data analysis (EDA), one-hot encoding for categorical features, and standard scaling for numerical features. Multiple models, including Logistic Regression and Random Forest, were trained and evaluated. The Random Forest model was further optimized using GridSearchCV to find the best hyperparameters, resulting in improved accuracy. The final tuned model is saved using pickle and deployed as an interactive web application with Streamlit for real-time predictions.

📅6/18/2025
⏱️
👤Machine Learning Engineer
completed
Project screenshot 1

Model Performance

80%
Accuracy
66%
Precision
50%
Recall
57%
F1 Score

Technologies Used

Pythonscikit-learnpandasnumpymatplotlibseabornpicklestreamlit

Features

Data preprocessing including handling missing values and data type conversion
One-hot encoding for categorical variables and feature scaling
Training and comparison of Logistic Regression and Random Forest models
Hyperparameter tuning using GridSearchCV for model optimization
Model evaluation using classification reports and confusion matrices
Feature importance analysis with Random Forest
Saving the trained model and features using pickle
Deployment as an interactive web application using Streamlit