A dataset for building the Einstein Discovery story contains 72 fields that are potentially relevant predictors.
Which approach is considered best practice to assess the top predictors in order to get to a meaningful and robust model?

Question

A dataset for building the Einstein Discovery story contains 72 fields that are potentially relevant predictors.

Which approach is considered best practice to assess the top predictors in order to get to a meaningful and robust model?

Bjorn-Rune Ness Lind · Accepted Answer

Build a story with a first set of predictors and assess which predictors are important to the story. Then drop the less important ones and add the predictors that were omitted in the first run and assess their impact.

Bjorn-Rune Ness Lind · Answer

This dataset is too big and cannot be used in Einstein Discovery. Request a new dataset with fewer predictors.

Bjorn-Rune Ness Lind · Answer

Build the story with all the predictors and indicate that Einstein Discovery should show the top predictors.

Bjorn-Rune Ness Lind · Answer

Go back to the data preparation and reduce the number of fields to less than 30 in order to produce a story.

Question list

List of questions

Question 1

(0)

Question 2

(0)

Question 3

(0)

Question 4

(0)

Question 5

(0)

Question 6

(0)

Question 7

(0)

Question 8

(0)

Question 9

(0)

Question 10

(0)

Related questions

Question 49 - Certified Tableau CRM and Einstein Discovery Consultant discussion

Suggested answer: D

0 comments