MENU
Case study
An ML engineer is developing a fraud detection model on AWS. The training dataset includes transaction logs, customer profiles, and tables from an on-premises MySQL database. The transaction logs and customer profiles are stored in Amazon S3.
The dataset has a class imbalance that affects the learning of the model's algorithm. Additionally, many of the features have interdependencies. The algorithm is not capturing all the desired underlying patterns in the data.
After the data is aggregated, the ML engineer must implement a solution to automatically detect anomalies in the data and to visualize the result.
Which solution will meet these requirements?
Suggested Answer: C
Get Premium MLA-C01 Questions
as Interactive Web-Based Practice Test or PDF
Real Exam Format
Detailed Explanations
Instant Access
Get Your Practice Test
30 comments · newest first
JU
Juan
10 months ago
Option A seems too basic for this kind of task.
NI
Nida
10 months ago
QuickSight is great for visualization, but I’m not sure about using it with AWS Batch.
HY
Hyun
10 months ago
I disagree, I believe option B would work better with Redshift Spectrum.
KI
Kiley
11 months ago
I think option C is the best choice for anomaly detection and visualization.
ET
Ettie
11 months ago
Wait, can SageMaker really handle both detection and visualization? Sounds too good to be true!
LI
Lindsey
11 months ago
I think AWS Batch is more for processing jobs rather than anomaly detection, but I'm not completely confident about that.
HO
Honey
11 months ago
I practiced a similar question where QuickSight was used for visualization, but I can't recall if it was paired with Redshift or something else.
MA
Mabelle
11 months ago
I'm not entirely sure, but I feel like Amazon Athena is more for querying data rather than detecting anomalies.
TR
Tricia
12 months ago
I remember we discussed class imbalance in our study group, and I think using SageMaker might help with that.
MA
Mariko
12 months ago
Option C with Amazon SageMaker Data Wrangler could be a good all-in-one solution, but I'll need to research how well it handles the specific challenges mentioned in the case study.
DE
Desiree
12 months ago
The data is spread across S3 and an on-premises MySQL database, so I'll need to consider how each option handles that kind of hybrid data setup.
LU
Luis
1 year ago
I'm leaning towards option B - using Amazon Redshift Spectrum to detect the anomalies and Amazon QuickSight for visualization. That seems like a robust and flexible approach.
JO
Joseph
1 year ago
Hmm, the class imbalance and feature interdependencies in the data make this a bit tricky. I'll need to think through how each of the options might handle those challenges.
AU
Audria
1 year ago
This seems like a straightforward case study, but I'll need to carefully read through the details to identify the best solution. The key requirements are automatically detecting anomalies and visualizing the results.
JA
Jaclyn
1 year ago
You know, as an ML engineer, I'm always looking for the path of least resistance. Option C seems to cover all the bases, and I don't have to worry about setting up a whole data warehouse. Gotta love that efficiency!
KA
Kattie
1 year ago
Definitely, efficiency is key in our line of work. Let's go with option C for this project.
EL
Ellsworth
1 year ago
I agree, using SageMaker Data Wrangler will save us a lot of time and effort. It's the most efficient option.
LA
Lashandra
1 year ago
Option C seems like the best choice. Amazon SageMaker Data Wrangler can handle both anomaly detection and visualization.
LO
Lorean
1 year ago
That's a good point, but I still think option C is more suitable for this specific case study.
DA
Dallas
1 year ago
Hold up, did someone say 'class imbalance'? That's my jam! Option C sounds like it has the tools to handle that, plus the visualization. I'm feeling confident about this one.
AR
Arlie
1 year ago
Hmm, I'm not sure. Option B with Redshift Spectrum and QuickSight could also work, but it might be overkill for this use case. I'd need to dig deeper into the requirements to make a more informed decision.
SH
Shawna
1 year ago
True, but I still think option D with AWS Batch and QuickSight could be a simpler solution for this case. It's worth considering.
GI
Gilma
1 year ago
I agree, but using Amazon Athena from option A might also be a good fit. It can handle large-scale data processing efficiently.
IZ
Izetta
1 year ago
I think option C with SageMaker Data Wrangler could be a good choice. It's designed for handling complex datasets.
VO
Von
1 year ago
I disagree, I believe option B is better as Amazon Redshift Spectrum can handle large datasets efficiently.
LO
Lorean
1 year ago
I think option C is the best choice because Amazon SageMaker Data Wrangler can handle the class imbalance and interdependencies.
MA
Ma
1 year ago
I think option C is the way to go. Amazon SageMaker Data Wrangler has all the tools to handle the data preprocessing and anomaly detection, plus the visualization capabilities to get the job done.
DO
Dominga
1 year ago
I would go with option B, using Amazon Redshift Spectrum and QuickSight seems like a solid choice.
CA
Carolann
1 year ago
I think option A could also work well, Amazon Athena is powerful for detecting anomalies.
VA
Vallie
1 year ago
I agree, option C seems like the most comprehensive solution for this case study.