Data Analytics · HR Analytics · Machine Learning

HR Analytics & Employee Attrition Prediction System

An end-to-end HR analytics project that analyzes employee attrition patterns and predicts turnover using SQL, Python, Power BI, machine learning, and a Streamlit app.

Business Problems

Employee attrition carries significant cost and operational impact. This project looks at the factors behind attrition and builds a predictive model to support earlier intervention.

  • Recruitment and training costs
  • Employee retention
  • Satisfaction and performance
  • Workload management
  • Salary-related attrition patterns
  • Workforce planning

Analytical Workflow

SQL Cleaning→ EDA→ Visualization→ Power BI→ ML Modeling→ Streamlit App

Key Analysis

  • SQL data cleaning and validation
  • Attrition distribution
  • Satisfaction analysis
  • Salary analysis
  • Department analysis
  • Employee tenure
  • Workload analysis
  • Correlation analysis
  • Attrition driver identification

Model Comparison — Reported Results

75.34%Logistic Regression Accuracy
53.17%LR Precision
72.18%LR Recall
95.13%Random Forest Accuracy
87.77%RF Precision
92.23%RF Recall

These are reported model evaluation results from the project. They should not be interpreted as guarantees of employee retention or as the sole basis for employment decisions.

Feature Importance (Reported)

  • Satisfaction Level
  • Time Spent at Company
  • Number of Projects
  • Average Monthly Hours
  • Last Evaluation

Technologies Used

PythonPandasNumPyMatplotlibSeaborn MySQLSQLPower BIScikit-learn Logistic RegressionRandom ForestStreamlit

Project Resources