QuestionQ26

VerifiedModeling, Analysis, and Outcomes

A data scientist is building a model to predict the result of a vote for a national mascot. The options are tigers and lions. The complete data set contains feedback from people representing 17 professions and 12 distinct locations. The following rank aggregation accounts for 80% of the data set:

Question Image

Which of the following is the most likely concern regarding the model's ability to predict the vote outcome?

Explanation

The data represented is concentrated in only a small subset of the 17 professions and 12 locations. Predicting preferences for professions or locations not represented in that concentrated sample requires extending the model beyond the observed data distribution, which is extrapolation and can produce unreliable predictions.

Community Discussion

No comments yet. Be the first to start the discussion!