Databricks Certified Professional Data Scientist Exam - Topic 4 Question 53 Discussion

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

Question #: 53
Topic #: 4

[All Databricks Certified Professional Data Scientist Questions]

Refer to the exhibit.

You are using K-means clustering to classify customer behavior for a large retailer. You need to determine the optimum number of customer groups. You plot the within-sum-of-squares (wss) data as shown in the exhibit. How many customer groups should you specify?

Show Suggested Answer

Suggested Answer: B

by Portia at Feb 11, 2024, 12:17 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

Benedict

4 months ago

2 groups might be too simplistic for customer behavior.

upvoted 0 times

...

Carolynn

5 months ago

Wait, are we sure about this? What if there's more nuance?

upvoted 0 times

...

Stevie

5 months ago

I thought it was 4 at first, but 3 makes sense.

upvoted 0 times

...

Arletta

5 months ago

Definitely agree, 3 seems optimal!

upvoted 0 times

...

Jame

5 months ago

Looks like the elbow point is at 3 groups.

upvoted 0 times

...

Jess

6 months ago

I think I read that you should look for a point where the wss starts to level off. That might be around 3 or 4, but I’m not confident.

upvoted 0 times

...

Owen

6 months ago

I practiced a similar question where the elbow was quite clear. If I recall correctly, it was around 2 or 3, but I can't remember the exact details.

upvoted 0 times

...

Lavonne

6 months ago

I'm not entirely sure, but I feel like the wss decreases significantly up to a certain point. Maybe 4 could be the right choice?

upvoted 0 times

...

Naomi

6 months ago

I remember we discussed the elbow method for determining the optimal number of clusters. I think it might be around 3 based on the graph.

upvoted 0 times

...

Katie

6 months ago

8 groups seems like overkill for this data. The plot shows the biggest drop-off is around 3 or 4, so I'd go with one of those options.

upvoted 0 times

...

Mayra

6 months ago

Based on the elbow plot, I'd say the best choice is 4 customer groups. The curve starts to flatten out after that point.

upvoted 0 times

...

Yuonne

6 months ago

Hmm, I'm not sure about this one. The plot doesn't have a super obvious elbow, so it's hard to tell. I might need to do some more analysis to decide.

upvoted 0 times

...

Tess

6 months ago

The elbow plot shows a clear bend at 3, so I think the optimum number of customer groups is 3.

upvoted 0 times

...

Shanda

6 months ago

Time to put my problem-solving skills to the test. I'll carefully analyze each of the answer choices and see which one makes the most logical sense.

upvoted 0 times

...

Jerilyn

6 months ago

I'm a bit confused by the dedupratio property mentioned in the options. I'll need to research that a bit more to understand how it works and how it can help us achieve the desired space savings.

upvoted 0 times

...

Rosio

6 months ago

Hmm, this is an interesting one. I think the key might be understanding how Visualforce controls interact with the CCRZ framework. I'll need to review that to figure out the best approach.

upvoted 0 times

...

Jina

6 months ago

I'm a bit confused by the wording of this question. I'll need to re-read it a few times to make sure I understand what they're asking for before selecting an answer.

upvoted 0 times

...

Halina

6 months ago

I feel like I saw something about Azure too, but AWS keeps coming to mind as the main one for Aviatrix.

upvoted 0 times

...

Isidra

6 months ago

I'm pretty lost on this. I know share capital is credit, but I'm not sure about the others—need to think this through more!

upvoted 0 times

...

Moon

11 months ago

Wait, is that a graph or a plate of spaghetti? I'm getting hungry just looking at it. But in all seriousness, I'd say 3 groups is the way to go. The elbow is as clear as day.

upvoted 0 times

Narcisa

10 months ago

I would go with 3 groups too, the graph clearly shows the elbow at that point.

upvoted 0 times

...

Lezlie

10 months ago

I think 3 groups makes sense based on the elbow point in the graph.

upvoted 0 times

...

Quiana

10 months ago

I agree, the elbow method is a great way to determine the optimal number of clusters.

upvoted 0 times

...

Marcos

11 months ago

2 groups? What is this, the stone age? This is the 21st century, people! We need at least 8 groups to truly understand our customers. Anything less is just lazy data science.

upvoted 0 times

...

Lucia

11 months ago

Hmm, I don't know, the 'elbow' looks a bit crooked to me. I'm going to go with 4 groups just to be safe. Can't be too careful with customer segmentation, you know.

upvoted 0 times

Ceola

10 months ago

User 4: I'm not sure, but I'll go with 4 groups too.

upvoted 0 times

...

Rochell

10 months ago

User 3: I see what you mean, but I think 4 groups might be safer.

upvoted 0 times

...

Emmanuel

10 months ago

User 2: I agree, I would go with 3 groups as well.

upvoted 0 times

...

Amira

10 months ago

User 1: I think the elbow method is showing 3 groups.

upvoted 0 times

...

Fallon

11 months ago

Wait, is this a trick question? The graph clearly shows the optimal number of groups is 8. Anything less and we're missing out on valuable customer insights!

upvoted 0 times