This project performs an in-depth exploratory data analysis (EDA) on an HR dataset to understand the factors that influence employee attrition. The goal is to uncover insights that HR teams can use to improve retention and reduce turnover.
To analyze employee attrition patterns using data analysis techniques and generate HR-driven insights on:
- Age, income, and attrition
- Job satisfaction, overtime, and promotions
- Work-life balance and relationship satisfaction
- Departmental and job role trends
- Python 3
- Jupyter Notebook (
.ipynb) - Pandas, NumPy
- Seaborn, Matplotlib
- Git, GitHub
| File | Description |
|---|---|
HR_Attrition_Analysis.ipynb |
Main notebook containing all EDA steps and visualizations |
hr_dataset.csv |
HR dataset used for analysis (ensure you're allowed to share) |
- Employees with poor work-life balance, lower relationship satisfaction, and frequent overtime are more likely to leave.
- Lower monthly income and longer times since last promotion are associated with higher attrition.
- Certain roles (like Sales Representatives) have higher exit rates.
- Most attrition happens in employees with 1–5 years of experience.
- Improve employee engagement through regular promotions and upskilling
- Monitor work-life balance and overtime load
- Focus on at-risk roles and departments
- Launch retention campaigns targeting younger and low-income employees
- Build machine learning models to predict attrition
- Create an interactive dashboard using Power BI or Streamlit
- Integrate HR KPIs into a live analytics portal
Puli Eswar
LinkedIn | GitHub
(Data Analyst | Python | EDA | Visualization)