Skip to content

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

After a customer-segmentation model is deployed by a Bengaluru e-commerce company, its predictions slowly worsen as buying behaviour changes after a festive season. Which lifecycle activity is mainly needed?

Ongoing monitoring and periodic retraining is needed. Customer behaviour changes cause model drift, so performance must be tracked after deployment and the model updated with recent data. Removing dashboards or printing reports does not restore predictive accuracy.

  1. AOngoing monitoring and periodic retraining of the modelCorrect
  2. BRestarting only data collection and discarding the model
  3. CRemoving visualisations from dashboards
  4. DConverting the model to a paper report

Explanation

Changing behaviour causes model drift, so deployed models must be monitored and retrained with fresh data. Discarding the model or altering dashboards does not fix degraded predictions, and a paper report does not restore accuracy.

Did you get it right without looking?

One question tells you little. A timed set on Data Analytics shows your real accuracy, how long you take and where you lose marks.

More Data Analytics questions