HR Dataset

The Kaggle HR Dataset, created by rhuebner, is one of the most popular open datasets for practicing human resources analytics. It contains detailed employee data including satisfaction levels, evaluation scores, number of projects, average monthly hours, time spent in the company, promotion history, and salary category.

Available on Kaggle, this dataset is excellent for building predictive models that help organizations understand and improve employee performance, engagement, and retention.

Key Features

• Records: Hundreds of employee profiles (depending on version).

• Variables:

  • Satisfaction level

  • Last evaluation score

  • Number of projects

  • Average monthly hours

  • Time spent in company (years)

  • Promotion in last 5 years (0/1)

  • Salary category (low, medium, high)

• Data Type: Mixed (numerical and categorical).

• Format: ZIP file.

Why This Dataset 

This dataset allows data scientists and HR analysts to explore how workplace conditions and employee metrics relate to retention, performance, and satisfaction. It’s ideal for projects that aim to:

1. Predict employee turnover (attrition).

2. Identify the most important factors influencing satisfaction.

3. Segment employees using clustering methods.

4. Detect patterns in promotion and salary data.

5. Build fair and transparent HR models with explainable AI techniques.

How to Use the Dataset

1. Download the ZIP file

2. Extract and load it into Python (Pandas) or R.

3. Explore the data using .describe() and .info().

4. Handle missing values and encode categorical features such as salary level.

5. Train classification models (Logistic Regression, Random Forest, XGBoost).

6. Evaluate using metrics like Accuracy, ROC-AUC, and Precision-Recall.

7. Visualize insights with Seaborn or Matplotlib (e.g., heatmaps, feature importance).

Possible Project Ideas

• Employee attrition prediction model.

• Promotion recommendation system.

• Satisfaction score prediction using regression models.

• Bias detection in salary and promotion decisions.

• Clustering employees based on workload and performance metrics.