Noga-66/Employee-Churn-Analysis

This project was developed as the final project for the Machine Learning training program provided by the National Telecommunication Institute (NTI), Egypt. It focuses on applying supervised learning algorithms specifically Decision Tree and Random Forest to address a practical business challenge in the field of Human Resources predicting employee # Employee-Churn-Analysis This project was developed as the final project for the Machine Learning training program provided by the National Telecommunication Institute (NTI), Egypt. It focuses on applying supervised learning algorithms specifically Decision Tree and Random Forest to address a practical business challenge in the field of Human Resources predicting employee churn By leveraging data-driven insights, the project aims to help organizations proactively identify employees at risk of leaving, allowing HR departments to take timely and informed retention actions. --- ## 📁 Dataset Overview – *Employee Turnover (Churn) Dataset* This dataset provides detailed information about **14,999 employees** and is designed to support **employee churn prediction** – the task of determining whether an employee is likely to leave the company or not. It is a typical **binary classification problem**, where the target variable (`quit`) indicates whether an employee **has left (1)** or **is still working (0)**. The dataset includes **organizational, behavioral, and performance-related features** that can be used to build predictive models. --- ## 🎯 Project Goal The goal of this project is to **build machine learning models** that help Human Resources (HR) teams proactively identify employees who are likely to leave, so the organization can take corrective actions. You will: * Build and train **Decision Tree** and **Random Forest** classifiers. * Evaluate model performance using metrics like accuracy and confusion matrix. * Improve model performance through **hyperparameter tuning**, focusing on: * `max_depth` * `min_samples_split` * `min_samples_leaf` --- ## 📊 Dataset Structure * **Number of Instances (Rows):** 14,999 * **Number of Features (Columns):** 10 (independent variables) + 1 target variable (`quit`) * **Missing Values:** None * **Target Variable:** `quit` (1 = employee left, 0 = employee stayed) --- ## 🧾 Feature Dictionary | Column Name | …