Employee-Turnover-and-HR-data-Exploration

Project Overview

The project aims to understand factors contributing to employee turnover and create a predictive model to identify employees at risk of leaving the company.

OSEMN Data Science Pipeline

1. Obtaining the Data:

   • Downloaded the dataset from Kaggle.
   • Imported the data into the working environment.

2. Scrubbing the Data:

   • Checked for missing values (dataset was clean).
   • Examined the dataset for readability and appropriate feature names.
   • Converted categorical features (department, salary) to numeric types.

3. Exploratory Data Analysis (EDA):

   • Conducted statistical overview and summary.
   • Explored correlations among features using a correlation matrix and heatmap.
   • Analyzed turnover patterns in relation to department, salary, promotion, years at the company, project count, evaluation, average monthly hours, etc.

4. Modeling the Data:

   • Split the data into training and testing sets.
   • Implemented various machine learning models (Logistic Regression, SVM, kNN, Random Forest).
   • Evaluated model performance using training and testing scores.

5. Interpreting the Data:

   • Summarized findings from EDA.
   • Highlighted trends related to turnover, satisfaction, salary, project count, and evaluations.
   • Raised questions for further consideration about the impact of losing employees and factors affecting satisfaction and turnover.

Conclusion and Questions

• Noted trends related to working hours, salary, promotion, and project count.
• Highlighted correlations between turnover, satisfaction, and salary.
• Posed questions about the impact of losing employees and factors influencing turnover and satisfaction.

Note: The code sections may need to be reformatted and executed in a Python environment for full functionality.