Exam Certified Tableau CRM and Einstein Discovery Consultant topic 1 question 50 discussion

I agree that out of the given options, the two that seems better are A & D.... but B should not be out of the game

C. Proceed with deployment if the correlation values are sufficient. This is not a recommended next step, as correlation values that are too low may indicate that the model is not accurately capturing the relationships between variables in the data, which could lead to poor predictive performance when the model is deployed. D. Edit the model accuracy settings and rerun the story. This is not a recommended next step as it is not clear what accuracy settings should be changed, and doing so may not necessarily improve the model's correlation values or predictive power

A. Identify additional data that may have a stronger relationship with the outcome variable. This is a valid next step as it could lead to the discovery of additional variables that have a stronger correlation with the outcome variable, and thus improve the model's predictive power. B. Rerun and update the story with a different algorithm. This is also a valid next step, as using a different algorithm may improve the model's ability to detect patterns and relationships in the data, potentially resulting in higher correlation values.

I am a bit puzzled by the word "correlation" here. Statistically speaking, correlation would be the "importance" column in data prep screen. R² are the global model accuracy/metrics

A and D seems to be correct! https://help.salesforce.com/s/articleView?id=sf.bi_edd_story_modify.htm&type=5 FYI. correlation values (r value) below 20% are considered very weak. This might indicate that option C is not a correct answer. https://www.dummies.com/article/academics-the-arts/math/statistics/how-to-interpret-a-correlation-coefficient-r-169792/

Question was on exam 21 August 2022.