Skip to content

audrey-ma/Employee-Attrition-Prediction

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

2 Commits
 
 

Repository files navigation

Employee-Attrition-Prediction

IBM HR Analytics Clustering Project

This project applies clustering techniques to IBM HR analytics data to identify patterns and insights that can guide decision-making in human resources management. The analysis leverages various machine learning methods to uncover employee trends and segments.

Table of Contents

  • Introduction
  • Objectives
  • Data Overview
  • Methods
  • Results and Insights
  • Model Performance
  • How to Use
  • Contributors

Introduction

Human resource analytics is crucial for understanding workforce dynamics and improving organizational outcomes. This project employs clustering algorithms to explore patterns in employee data, focusing on areas like attrition, performance, and engagement.

Objectives

  • Identify key employee clusters based on demographic and behavioral attributes.
  • Provide actionable insights to improve employee retention and satisfaction.
  • Assess and visualize the performance of various clustering models.

Data Overview

The dataset used in this project includes anonymized HR metrics, such as:

  • Demographic Information
  • Job Role and Satisfaction
  • Attrition Rate
  • Performance Scores

Methods

The following techniques were employed:

  • Data Preprocessing: Handling missing values, scaling features, and encoding categorical variables.
  • Clustering Algorithms:
    • K-Means
    • Hierarchical Clustering
    • DBSCAN
  • Evaluation Metrics: Silhouette Score, Davies-Bouldin Index, and more.
  • Visualization: PCA and t-SNE for dimensionality reduction.

Results and Insights

Key findings include:

  • [Add high-level insights here, e.g., "Employees with higher engagement scores were less likely to leave the company."]
  • Visual representations of clusters revealed distinct employee groups.

Model Performance

Visualization of Results

image

image

image

image

image

  • Add performance graphs and tables here, such as:
    • A bar chart comparing silhouette scores across models.
    • A scatter plot showing clustered data in reduced dimensions.
    • A table summarizing evaluation metrics for each algorithm.

How to Use

  1. Clone the repository:

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published