Employee Attrition Prediction System Using Machine Learning

Employee Attrition Prediction System

Overview

This project is a professional web application that helps companies predict which employees might leave the organization. It uses machine learning and data analysis to give HR teams helpful insights for improving employee retention.The system includes easy-to-understand data visuals, prediction tools, and a clean web interface. It’s perfect for HR professionals, students, or data analysts working on real-world problems.

Project Summary

Download New Real Time Projects :-Click here

Available Features

• Predictive modeling for employee attrition
• Interactive web application built with Streamlit
• Data analysis and visualizations using Jupyter Notebook
• Insightful dashboard view for management reporting
• Key employee-related factors such as age, salary, job level, and training
• Pre- and post-prediction interface screenshots
• Machine learning model packaged for reuse
• Power BI integration for visual storytelling

Dataset Description

The dataset used in this project includes the following fields:

• Age – Employee’s age category
• Work-Life Balance – A score between 1 and 4 representing work-life satisfaction
• Monthly Salary – Monthly earnings
• Gender – Male or Female
• Training Sessions – Number of training programs attended in the previous year
• Job Level – Employee level: Entry, Mid, or Senior
• Attrition – Indicates whether the employee has left the organization

The data is stored in CSV format and does not rely on any external database system.

Key Insights

Age and Attrition

• Employees aged between 26 to 35 years are more likely to leave the organization compared to other age groups.

Work-Life Balance

• 65% of employees enjoy a healthy work-life balance, while 35% report challenges that may contribute to attrition.

Salary Differences

• Employees who left the company earned an average of $4,787, whereas those who stayed earned an average of $6,833.

Training Gaps

• Entry-level employees who left the organization had limited access to training, with 97% having attended fewer than 5 sessions in the prior year.

Recommendations

Based on the analysis, here are data-driven recommendations:

1. Retention Programs for Younger Employees
   Develop mentorship, growth paths, and targeted engagement strategies for the 26–35 age group.
2. Improve Work-Life Balance
   Introduce policies such as flexible schedules and mental health programs to address the 35% experiencing imbalance.
3. Salary Adjustments
   Benchmark compensation against industry standards and ensure competitive packages to retain top talent.
4. Training and Development
   Increase the number and quality of training sessions, particularly for entry-level staff, to improve skills and job satisfaction.

Files Included

All essential components required to understand, run, and customize the project are provided:

• data/attrition_data.csv – Raw dataset
• data/train_data.csv – Cleaned and processed training data
• notebooks/Employee_Attrition.ipynb – Jupyter notebook with data analysis and model building
• app/app.py – Streamlit app script
• model_and_key_components.pkl – Pre-trained model and pipeline
• requirements.txt – Python dependencies
• screenshots/ – Pre- and post-prediction visuals

Technical Stack

• Programming Language: Python
• Libraries: pandas, numpy, matplotlib, seaborn, scikit-learn
• Visualization: Power BI
• Web Framework: Streamlit
• Data Storage: CSV files
• Environment: Jupyter Notebook for exploration, Streamlit for deployment

We have projects Available in all languages:–Click Here