CompTIA DY0-001 Exam - Topic 1 Question 5 Discussion

Actual exam question for CompTIA's DY0-001 exam

Question #: 5
Topic #: 1

A data scientist is clustering a data set but does not want to specify the number of clusters present. Which of the following algorithms should the data scientist use?

ADBSCAN

Bk-nearest neighbors

Ck-means

DLogistic regression

Show Suggested Answer

Suggested Answer: A

DBSCAN discovers clusters based on density without requiring you to predefine the number of clusters, automatically finding arbitrarily shaped groups and identifying noise points.

by Carey at Jun 12, 2025, 03:53 AM

Limited Time Offer

25%

Off

Get Premium DY0-001 Questions as Interactive Web-Based Practice Test or PDF

Contribute your Thoughts:

Submit Cancel

Chandra

2 months ago

I thought k-nearest neighbors could work too, but maybe not!

upvoted 0 times

...

Jeannetta

2 months ago

K-means is not the right choice here, for sure.

upvoted 0 times

...

Pete

3 months ago

Wait, can DBSCAN really handle that well?

upvoted 0 times

...

Skye

3 months ago

Totally agree, no need to set clusters upfront.

upvoted 0 times

...

Jarod

3 months ago

DBSCAN is the way to go for that!

upvoted 0 times

...

Venita

3 months ago

I keep mixing up k-nearest neighbors and k-means, but I think k-nearest neighbors is more about classification, not clustering.

upvoted 0 times

...

Mitsue

4 months ago

I feel like we practiced a question similar to this, and I think it was about density-based clustering, which makes me lean towards DBSCAN.

upvoted 0 times

...

Lorean

4 months ago

I'm not entirely sure, but I think k-means requires you to set the number of clusters beforehand, so it can't be the answer.

upvoted 0 times

...

Pearly

4 months ago

I remember that DBSCAN is good for finding clusters without needing to specify the number of clusters, so I think that might be the right choice.

upvoted 0 times

...

Rodolfo

4 months ago

Logistic regression? That doesn't seem right for a clustering problem. I'm leaning towards DBSCAN or maybe k-nearest neighbors, but I'll have to review the differences between those approaches.

upvoted 0 times

...

Cassandra

4 months ago

DBSCAN sounds like the right choice since the question specifically says the data scientist doesn't want to specify the number of clusters. I feel confident about this one.

upvoted 0 times

...

Leota

5 months ago

Hmm, I'm a bit unsure about this one. I know k-means requires specifying the number of clusters, but I'm not as familiar with the other algorithms. I'll have to think this through carefully.

upvoted 0 times

...

Novella

5 months ago

I'm pretty sure DBSCAN is the algorithm we want here since it can automatically determine the number of clusters without needing to specify it.

upvoted 0 times

...

Desiree

8 months ago

As a data scientist, I'd rather not 'k-means' my way through this problem. DBSCAN is the way to 'cluster' the competition.

upvoted 0 times

Kaycee

6 months ago

I agree, DBSCAN is great for clustering when you don't want to specify the number of clusters.

upvoted 0 times

...

Leonie

7 months ago

DBSCAN is definitely the way to go for this problem.

upvoted 0 times

...

Sherell

8 months ago

But DBSCAN is specifically designed for clustering without specifying the number of clusters.

upvoted 0 times

...

Alton

8 months ago

Hmm, logistic regression? That's for classification, not clustering. Clearly, the data scientist needs to brush up on their unsupervised learning algorithms.

upvoted 0 times