Skip to content

CS Professional · Artificial Intelligence, Data Analytics and Cyber Security - Laws and Practice · Artificial Intelligence - Introduction and Basics

A Chennai firm finds that its deployed demand-forecasting model grows less accurate over months because customer behaviour has changed from the patterns in its training data. What is this phenomenon called, and what lifecycle stage addresses it?

This is model drift: the live data has shifted from the training data, so predictions degrade. It is handled in the post-deployment monitoring and maintenance stage by tracking performance and retraining or updating the model with current data.

  1. AData anonymisation, addressed at the data collection stage
  2. BModel drift, addressed through monitoring and retraining after deploymentCorrect
  3. COverclocking, addressed at the hardware design stage
  4. DFeature scaling, addressed at the problem definition stage

Explanation

When real-world data changes so that a trained model's predictions degrade, this is model (or data/concept) drift. It is handled in the monitoring and maintenance stage by tracking performance and retraining or updating the model. The other options name unrelated ideas.

Did you get it right without looking?

One question tells you little. A timed set on Artificial Intelligence - Introduction and Basics shows your real accuracy, how long you take and where you lose marks.

Practise Artificial Intelligence - Introduction and BasicsFree diagnostic mock

More Artificial Intelligence - Introduction and Basics questions