Synthetic Employee Churn Records for Prediction Modeling

200,000 synthetic employee records are provided for churn prediction analysis. The dataset is designed to model customer or employee departure from a service. Its creator and temporal coverage are not specified.

Use Cases

• Predict churn likelihood using features like tenure, salary band, and department.
• Analyze the correlation between employee satisfaction scores and churn status.
• Build a classification model to identify high-risk employees based on synthetic performance metrics.
• Test feature engineering techniques on synthetic demographic and employment history data.

Strengths

• Contains 200,000 records, providing a substantial sample for model training.
• Synthetic nature allows for privacy-safe experimentation without real personal data.

Limitations

• Data is synthetic, limiting direct applicability to real-world business scenarios.
• The absence of real-world validation may obscure hidden biases or unrealistic feature relationships.
• Specific column definitions and data distributions are unknown, complicating model interpretation.

Provenance

Source

Kaggle

Collection Method

Synthetically generated.