Workforce Attrition Patterns and Risk Hotspot Analysis using Machine Learning

Read full text Summarise with VIRAT
Authors: Kunal Kashyap
Cited by: 0

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

This project presents a comprehensive analysis of workforce attrition patterns and identifies high-risk employee segments using statistical methods and machine learning techniques. The study focuses on understanding attrition behavior across departments, job roles, demographics, tenure, and workload factors such as overtime and business travel. Key methodologies used in this project include:- Exploratory Data Analysis (EDA)- Feature Engineering (AgeGroup, TenureGroup, IncomeCategory, etc.)- Hypothesis Testing (Chi-Square and T-Test)- Machine Learning Models (Logistic Regression, Random Forest, XGBoost)- Hyperparameter tuning using Optuna The final model achieved an ROC-AUC score of approximately 0.80, demonstrating strong predictive performance. An interactive Streamlit dashboard was developed to provide real-time insights, attrition risk prediction, and workforce analytics for HR decision-making. Key findings include:- Higher attrition in sales and high-pressure roles- Significant impact of overtime on employee exits- Early-career employees are more likely to leave- Job satisfaction and work-life balance are critical drivers This project demonstrates how data science can be applied to solve real-world HR problems and enable proactive workforce management.

Fields of Study & Subjects

AI and HR Technologies Customer churn and segmentation Employee Performance and Management

Publication details

Journal / source
Zenodo (CERN European Organization for Nuclear Research)
Publication date
21 March 2026
Language
EN