🧠 Employee Attrition Analysis & Prediction
This project aims to help HR professionals predict and reduce employee attrition using data-driven insights and machine learning models.
🔍 Project Overview
- Analyzed HR data to identify key drivers of attrition
- Built and compared classification models (Logistic Regression & Decision Tree)
- Predicted probability of attrition for each employee
- Exported a list of high-risk employees (
Attrition_Prob > 0.5) - Built an interactive dashboard to help HR understand who’s at risk and why
🧪 Technologies Used
- Python: pandas, matplotlib, seaborn, scikit-learn
- Jupyter Notebook: EDA, modeling, exporting data
- Power BI: Dashboard and interactive filtering
- CSV: Input and output datasets
📊 Dashboard Features
Built using Power BI to help HR teams make informed decisions:
| Section | Description |
|---|---|
| ✅ KPIs | Total at-risk employees, average attrition probability, avg. income |
| 📈 Charts | No. of Employees By Risk Level |
| 📍 Scatter Plot | MonthlyIncome vs Attrition_Prob |
| 📌 Filters | Department |
| 🔴 Table | Top high-risk employees |
📌 Key Takeaways
- Salary is not the only factor—OverTime and JobSatisfaction play critical roles
- Some high-income employees still showed high attrition risk
- Actionable recommendations (e.g., reduce workload, improve satisfaction)