Deal of The Day! Hurry Up, Grab the Special Discount - Save 25% - Ends In 00:00:00 Coupon code: SAVE25
Welcome to Pass4Success

- Free Preparation Discussions
Mail Us support@pass4success.com
Location US
Mob_nav_barsMENU

Amazon MLS-C01 Exam - Topic 3 Question 118 Discussion

[Modeling]A Data Scientist is building a model to predict customer churn using a dataset of 100 continuous numericalfeatures. The Marketing team has not provided any insight about which features are relevant for churnprediction. The Marketing team wants to interpret the model and see the direct impact of relevant features onthe model outcome. While training a logistic regression model, the Data Scientist observes that there is a widegap between the training and validation set accuracy.Which methods can the Data Scientist use to improve the model performance and satisfy the Marketing team'sneeds? (Choose two.)
A) Add L1 regularization to the classifier and C) Perform recursive feature elimination
B) Add features to the dataset
D) Perform t-distributed stochastic neighbor embedding (t-SNE)
E) Perform linear discriminant analysis

Amazon MLS-C01 Exam - Topic 3 Question 118 Discussion

Actual exam question for Amazon's MLS-C01 exam
Question #: 118
Topic #: 3
[All MLS-C01 Questions]

[Modeling]

A Data Scientist is building a model to predict customer churn using a dataset of 100 continuous numerical

features. The Marketing team has not provided any insight about which features are relevant for churn

prediction. The Marketing team wants to interpret the model and see the direct impact of relevant features on

the model outcome. While training a logistic regression model, the Data Scientist observes that there is a wide

gap between the training and validation set accuracy.

Which methods can the Data Scientist use to improve the model performance and satisfy the Marketing team's

needs? (Choose two.)

Show Suggested Answer Hide Answer
Suggested Answer: A, C

by Gearldine at Jul 09, 2025, 10:14 PM

Limited Time Offer

25%

Off

Get Premium MLS-C01 Questions as Interactive Web-Based Practice Test or PDF

Contribute your Thoughts:

0/2000 characters
Submit Cancel
Pearlene
6 months ago
Adding more features might not be the best move, though.
upvoted 0 times
...
Kelvin
6 months ago
I think recursive feature elimination is a solid choice too.
upvoted 0 times
...
Bev
6 months ago
Linear discriminant analysis could provide some insights as well!
upvoted 0 times
...
Sarina
6 months ago
Wait, can t-SNE really help with this? Seems off.
upvoted 0 times
...
Lashawn
6 months ago
L1 regularization can help reduce overfitting!
upvoted 0 times
...
Laurel
7 months ago
I recall a practice question where we used t-SNE for visualization, but I'm not sure it would directly improve model performance for this case.
upvoted 0 times
...
Latosha
7 months ago
I feel like adding more features could be risky, especially if we don't know which ones are relevant. It might just complicate things further.
upvoted 0 times
...
Carma
7 months ago
I'm not entirely sure, but I think performing recursive feature elimination might help identify the most relevant features for churn prediction.
upvoted 0 times
...
Daryl
7 months ago
I remember we discussed regularization techniques in class, so I think adding L1 regularization could help with overfitting.
upvoted 0 times
...
Madelyn
7 months ago
I've got this! The wide gap between training and validation accuracy suggests overfitting, so I'd definitely start with L1 regularization. And to satisfy the Marketing team's needs, recursive feature elimination is the way to go. Easy peasy!
upvoted 0 times
...
Nicolette
8 months ago
Okay, this is a tricky one. I think the key is to balance model performance and interpretability. Adding features could improve the model, but that might make it harder for the Marketing team to understand. I'd probably go with a combination of recursive feature elimination and linear discriminant analysis to identify the most relevant features.
upvoted 0 times
...
Tyisha
8 months ago
Hmm, I'm a bit unsure about this one. Adding L1 regularization could help with feature selection, but I'm not sure if that's the best approach here. Maybe I should also consider performing t-SNE to visualize the data and get a better understanding of the feature space.
upvoted 0 times
...
Evette
8 months ago
This looks like a classic feature selection problem. I'd start by trying recursive feature elimination to identify the most relevant features for the model.
upvoted 0 times
...
Charisse
10 months ago
Wait, linear discriminant analysis (option E)? I thought this was a logistic regression problem. That's like using a sledgehammer to crack a nut. Let's keep it simple, folks!
upvoted 0 times
...
Odelia
10 months ago
Hmm, I'm not sure t-SNE (option D) is the best fit for this problem. Isn't that more for visualization and dimensionality reduction? I think the Data Scientist should stick to methods that are more directly related to feature selection and model interpretation.
upvoted 0 times
...
Darrin
10 months ago
Option A, adding L1 regularization, could also be helpful in this case. It can help with feature selection and potentially improve the model's performance.
upvoted 0 times
Veronika
8 months ago
A: I think adding L1 regularization could help with feature selection.
upvoted 0 times
...
...
Jacob
10 months ago
Adding more features to the dataset might also be a good idea to improve accuracy.
upvoted 0 times
...
Gwenn
10 months ago
We could also try performing recursive feature elimination to see if that helps.
upvoted 0 times
...
Luann
11 months ago
Adding more features (option B) could be a good idea, but I'm not sure if that's the best approach here. The Marketing team wants to interpret the model, so focusing on feature selection might be more appropriate.
upvoted 0 times
Octavio
10 months ago
B: Adding L1 regularization to the classifier (option A) could also be a good method to address the gap between training and validation set accuracy.
upvoted 0 times
...
Arlette
10 months ago
A: I think performing recursive feature elimination (option C) could help improve the model performance and satisfy the Marketing team's needs.
upvoted 0 times
...
...
Jacob
11 months ago
I agree with Cletus, that could help improve the model performance.
upvoted 0 times
...
Cletus
11 months ago
I think we should add L1 regularization to the classifier.
upvoted 0 times
...
Dorothy
11 months ago
I think option C is the way to go. Recursive feature elimination can help the Data Scientist identify the most relevant features for churn prediction and satisfy the Marketing team's needs.
upvoted 0 times
Shalon
10 months ago
A: True, combining both options C and B could provide a more robust model for predicting customer churn.
upvoted 0 times
...
Pearly
10 months ago
B: Adding features to the dataset could also help in closing the gap between training and validation set accuracy.
upvoted 0 times
...
Enola
11 months ago
A: I agree, option C seems like a good choice to improve the model performance.
upvoted 0 times
...
...

Save Cancel