You built and manage a production system that is responsible for predicting sales numbers. Model accuracy is crucial, because the production model is required to keep up with market changes. Since…

Question

Options

• APoor data quality
• BLack of model retraining
• CToo few layers in the model for capturing information
• DIncorrect data split ratio during model training, evaluation, validation, and test

• A
  10% (4)
• B
  86% (36)
• C
  2% (1)
• D
  2% (1)

Explanation

This is a classic example of model staleness caused by data drift or concept drift. The model was trained on historical data that reflected past market conditions. Over time, market dynamics change - consumer behavior, pricing patterns, seasonal trends, and economic conditions all evolve - causing the real-world data distribution to diverge from what the model learned during training. Since the model itself has not been updated, its predictions become progressively less accurate. The solution is to regularly retrain the model on fresh data to keep it aligned with current market conditions. Poor data quality (A) would typically cause immediate issues rather than a steady decline. Architectural issues (C) or train/test split issues (D) would have been apparent at initial deployment, not as a gradual decline post-deployment.

No community discussion yet for this question.