Data Governance
Machine Learning
Changes in the distribution of input features over time while the relationship between features and target remains constant.
Detailed Explanation
Data drift occurs when the statistical properties of input features change over time, impacting the performance of machine learning models. Although the relationship between features and the target variable remains unchanged, the shift in data distribution can lead to inaccurate predictions and reduced model effectiveness, necessitating ongoing monitoring and potential model updates to maintain optimal performance.
Use Cases
- •Detect data drift to trigger model retraining and maintain prediction accuracy in evolving data environments.
Related Terms
Other terms in the Machine Learning category