Skip to content

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

A Pune fintech's credit-scoring model scores 99% accuracy on its training data but only 68% on fresh applicants. Which diagnosis best fits this pattern?

The pattern indicates overfitting. The model has memorised the training data, including its noise, so it scores very high there but generalises poorly to new applicants. Underfitting would instead produce weak results on both the training data and unseen data.

  1. AUnderfitting, because the model is too simple to capture patterns
  2. BOverfitting, because the model has memorised training data including noiseCorrect
  3. CData masking, because identifiers were hidden
  4. DConcept encryption, because the dataset was secured

Explanation

A large gap between very high training accuracy and much lower accuracy on unseen data signals overfitting: the model has learned noise specific to the training set. Underfitting would show poor performance on both training and new data.

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