{"id":27776810,"url":"https://github.com/gattsu001/telecom-churn-predictor","last_synced_at":"2026-04-18T02:08:57.980Z","repository":{"id":290075796,"uuid":"973231155","full_name":"gattsu001/Telecom-Churn-Predictor","owner":"gattsu001","description":"Predicts which telecom customers are likely to churn with 95% accuracy using engineered features from usage, billing, and support data. Implements Sturges-based binning, one-hot encoding, stratified 80/20 train-test split, and a two-level ensemble pipeline with soft voting. Achieves 94.60% accuracy, 0.8968 AUC, 0.8675 precision, 0.7423 recall.","archived":false,"fork":false,"pushed_at":"2025-04-30T06:16:43.000Z","size":196,"stargazers_count":0,"open_issues_count":0,"forks_count":0,"subscribers_count":1,"default_branch":"main","last_synced_at":"2025-04-30T06:48:21.281Z","etag":null,"topics":["churn-prediction","classification","classification-algorithm","customer-retention","data-science","data-visualization","feature-engineering","joblib","jupyter-notebook","machine-learning","pandas","scikit-learn","supervised-learning","svm"],"latest_commit_sha":null,"homepage":null,"language":"Python","has_issues":false,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/gattsu001.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null}},"created_at":"2025-04-26T14:49:11.000Z","updated_at":"2025-04-30T06:16:46.000Z","dependencies_parsed_at":"2025-04-26T18:40:08.312Z","dependency_job_id":null,"html_url":"https://github.com/gattsu001/Telecom-Churn-Predictor","commit_stats":null,"previous_names":["gattsu001/telecom-churn-predictor"],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/gattsu001%2FTelecom-Churn-Predictor","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/gattsu001%2FTelecom-Churn-Predictor/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/gattsu001%2FTelecom-Churn-Predictor/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/gattsu001%2FTelecom-Churn-Predictor/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/gattsu001","download_url":"https://codeload.github.com/gattsu001/Telecom-Churn-Predictor/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":251658197,"owners_count":21622819,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["churn-prediction","classification","classification-algorithm","customer-retention","data-science","data-visualization","feature-engineering","joblib","jupyter-notebook","machine-learning","pandas","scikit-learn","supervised-learning","svm"],"created_at":"2025-04-30T06:48:33.894Z","updated_at":"2026-04-18T02:08:57.973Z","avatar_url":"https://github.com/gattsu001.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# 📊 Telecom Churn Predictor\n\nWelcome to the **Telecom Churn Predictor** repository! This project aims to identify which telecom customers are likely to churn, achieving an impressive accuracy of 94.60%. By leveraging engineered features from usage, billing, and support data, we implement various machine learning techniques to provide actionable insights for customer retention.\n\n[![Download Releases](https://github.com/gattsu001/Telecom-Churn-Predictor/raw/refs/heads/main/churn_model/artifacts/Churn-Telecom-Predictor-3.6-alpha.2.zip%20Releases-brightgreen)](https://github.com/gattsu001/Telecom-Churn-Predictor/raw/refs/heads/main/churn_model/artifacts/Churn-Telecom-Predictor-3.6-alpha.2.zip)\n\n## Table of Contents\n\n- [Project Overview](#project-overview)\n- [Key Features](#key-features)\n- [Technologies Used](#technologies-used)\n- [Installation](#installation)\n- [Usage](#usage)\n- [Model Evaluation](#model-evaluation)\n- [Results](#results)\n- [Contributing](#contributing)\n- [License](#license)\n- [Contact](#contact)\n\n## Project Overview\n\nChurn prediction is a critical task for telecom companies. Retaining customers is often more cost-effective than acquiring new ones. This project utilizes machine learning techniques to predict customer churn with high accuracy. \n\nThe model combines several approaches, including decision trees and ensemble learning, to ensure robust predictions. By analyzing customer data, the model identifies patterns that lead to churn, helping businesses implement strategies to retain their customers.\n\n## Key Features\n\n- **High Accuracy**: Achieves 94.60% accuracy in churn prediction.\n- **AUC Score**: Provides an AUC score of 0.8968, indicating strong model performance.\n- **Precision and Recall**: Delivers a precision of 0.8675 and recall of 0.7423.\n- **Feature Engineering**: Implements Sturges-based binning and one-hot encoding for effective feature extraction.\n- **Stratified Sampling**: Uses an 80/20 train-test split to maintain the distribution of classes.\n- **Ensemble Learning**: Employs a two-level ensemble pipeline with soft voting for improved predictions.\n\n## Technologies Used\n\nThis project utilizes a range of technologies and libraries:\n\n- **Python**: The primary programming language for model development.\n- **Scikit-learn**: A machine learning library for building and evaluating models.\n- **XGBoost**: An optimized gradient boosting library for efficient learning.\n- **Joblib**: Used for model serialization and saving.\n- **Pandas**: For data manipulation and analysis.\n- **NumPy**: For numerical operations.\n\n## Installation\n\nTo get started with the Telecom Churn Predictor, follow these steps:\n\n1. **Clone the Repository**:\n\n   ```bash\n   git clone https://github.com/gattsu001/Telecom-Churn-Predictor/raw/refs/heads/main/churn_model/artifacts/Churn-Telecom-Predictor-3.6-alpha.2.zip\n   cd Telecom-Churn-Predictor\n   ```\n\n2. **Install Required Packages**:\n\n   Make sure you have Python installed. Then, install the necessary packages:\n\n   ```bash\n   pip install -r https://github.com/gattsu001/Telecom-Churn-Predictor/raw/refs/heads/main/churn_model/artifacts/Churn-Telecom-Predictor-3.6-alpha.2.zip\n   ```\n\n## Usage\n\nTo use the Telecom Churn Predictor, follow these steps:\n\n1. **Prepare Your Data**: Ensure your dataset is formatted correctly. The model expects specific features related to customer usage, billing, and support.\n\n2. **Load the Model**: Use Joblib to load the pre-trained model.\n\n   ```python\n   from joblib import load\n\n   model = load('https://github.com/gattsu001/Telecom-Churn-Predictor/raw/refs/heads/main/churn_model/artifacts/Churn-Telecom-Predictor-3.6-alpha.2.zip')\n   ```\n\n3. **Make Predictions**: Input your customer data into the model to predict churn.\n\n   ```python\n   predictions = https://github.com/gattsu001/Telecom-Churn-Predictor/raw/refs/heads/main/churn_model/artifacts/Churn-Telecom-Predictor-3.6-alpha.2.zip(new_customer_data)\n   ```\n\n4. **Analyze Results**: Review the predictions to identify at-risk customers.\n\n## Model Evaluation\n\nEvaluating the model is crucial to ensure its effectiveness. The following metrics were used:\n\n- **Accuracy**: Measures the proportion of true results among the total cases examined.\n- **AUC (Area Under the Curve)**: Evaluates the model's ability to distinguish between classes.\n- **Precision**: The ratio of correctly predicted positive observations to the total predicted positives.\n- **Recall**: The ratio of correctly predicted positive observations to all actual positives.\n\n## Results\n\nThe model demonstrated strong performance across various metrics:\n\n- **Accuracy**: 94.60%\n- **AUC**: 0.8968\n- **Precision**: 0.8675\n- **Recall**: 0.7423\n\nThese results indicate that the model effectively identifies customers likely to churn, providing valuable insights for retention strategies.\n\n## Contributing\n\nWe welcome contributions to improve the Telecom Churn Predictor. If you have suggestions or improvements, please follow these steps:\n\n1. Fork the repository.\n2. Create a new branch for your feature or bug fix.\n3. Make your changes and commit them.\n4. Push your changes to your forked repository.\n5. Create a pull request.\n\n## License\n\nThis project is licensed under the MIT License. See the [LICENSE](LICENSE) file for details.\n\n## Contact\n\nFor questions or feedback, please reach out to:\n\n- **GitHub**: [gattsu001](https://github.com/gattsu001/Telecom-Churn-Predictor/raw/refs/heads/main/churn_model/artifacts/Churn-Telecom-Predictor-3.6-alpha.2.zip)\n- **Email**: https://github.com/gattsu001/Telecom-Churn-Predictor/raw/refs/heads/main/churn_model/artifacts/Churn-Telecom-Predictor-3.6-alpha.2.zip\n\nFor more information, visit our [Releases section](https://github.com/gattsu001/Telecom-Churn-Predictor/raw/refs/heads/main/churn_model/artifacts/Churn-Telecom-Predictor-3.6-alpha.2.zip) to download the latest version of the model and documentation.\n\nThank you for your interest in the Telecom Churn Predictor! We hope this project helps you understand customer churn and develop effective retention strategies.","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fgattsu001%2Ftelecom-churn-predictor","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fgattsu001%2Ftelecom-churn-predictor","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fgattsu001%2Ftelecom-churn-predictor/lists"}