Databricks Certified Professional Data Scientist Exam - Topic 3 Question 9 Discussion

Actual exam question for Databricks's Databricks Certified Professional Data Scientist exam

Question #: 9
Topic #: 3

[All Databricks Certified Professional Data Scientist Questions]

You have data of 10.000 people who make the purchasing from a specific grocery store. You also have their income detail in the dat

a. You have created 5 clusters using this data. But in one of the cluster you see that only 30 people are falling as below 30, 2400, 2600, 2700, 2270 etc."

What would you do in this case?

AYou will be increasing number of clusters.

BYou will be decreasing the number of clusters.

CYou will remove that 30 people from dataset

DYou will be multiplying standard deviation with the 100
Decreasing the number of clusters will help in adjusting this outlier cluster to get adjusted in another cluster.

Show Suggested Answer

Suggested Answer: B

by Pamella at May 06, 2022, 02:05 PM

Limited Time Offer

25%

Off

Get Premium Databricks Certified Professional Data Scientist Questions as Interactive Web-Based Practice Test or PDF

Contribute your Thoughts:

Submit Cancel

Margurite

4 months ago

Totally agree, more clusters could give better insights!

upvoted 0 times

...

Ricarda

4 months ago

Decreasing clusters might just mask the issue.

upvoted 0 times

...

Geraldine

4 months ago

Wait, why would you remove those 30 people?

upvoted 0 times

...

Amira

4 months ago

I’d definitely consider increasing the clusters.

upvoted 0 times

...

Candra

5 months ago

Seems like a classic outlier situation.

upvoted 0 times

...

Belen

5 months ago

Multiplying the standard deviation sounds a bit odd to me, but I guess it could be a way to adjust for the outlier.

upvoted 0 times

...

Skye

5 months ago

I feel like decreasing the number of clusters could help, but I'm not confident if that would really solve the issue with the outlier.

upvoted 0 times

...

Johnathon

5 months ago

I remember a practice question where we had to deal with outliers, and I think removing those 30 people might be too extreme.

upvoted 0 times

...

Viva

5 months ago

I'm not entirely sure, but I think increasing the number of clusters could help capture the diversity in the data better.

upvoted 0 times

...

Heike

5 months ago

I'm not totally confident about this one. I think C is the best option, but I'm not 100% sure. I'll make a note to review mount units and systemd in more detail before the exam.

upvoted 0 times

...

Abel

5 months ago

I remember studying how credit card statements could reveal transaction patterns, but I'm not sure if they directly indicate skimming.

upvoted 0 times

...

Kenneth

5 months ago

I've got a strategy for this - I'll carefully read through the answer choices and think about the typical OSPF to BGP redistribution settings.

upvoted 0 times

...

Asuncion

5 months ago

Hmm, I'm not entirely sure about the details of BIG IP's web server monitoring. I'll need to think this through carefully and consider the different options.

upvoted 0 times

...