Free Amazon MLS-C01 Actual Exam Questions - Question 15 Discussion
A Machine Learning Specialist is designing a system for improving sales for a company. The objective
is to use the large amount of information the company has on users' behavior and product
preferences to predict which products users would like based on the users' similarity to other users.
What should the Specialist do to meet this objective?
B, since the question highlights using user similarity to predict preferences, collaborative filtering fits best. Content-based (A) focuses more on item features, so it’s less relevant here.
It’s B for me because the focus is on leveraging similarities between users to make recommendations. Content-based filtering (A) mainly uses product features instead of user-user relationships, so it doesn’t fit as well here. Model-based filtering (C) isn’t a widely recognized term in this context, and combinative (D) seems like a mix without clear definition. Collaborative filtering directly captures user similarity to suggest products others with similar tastes liked, which matches the question’s goal perfectly.
Sounds like B is the right pick since it’s specifically about using user similarity to recommend products, which is what collaborative filtering does. Content-based (A) usually focuses on the product features themselves rather than comparing users, so that seems less fitting here.
B definitely, since it’s all about predicting based on user similarities, classic collaborative filtering.
A imo, because the question highlights using user behavior and product preferences, which sounds more like content-based filtering that leverages product attributes and user profiles directly. Collaborative filtering (B) is good for user similarity but usually relies on interaction data rather than explicit product info. Also, option D doesn’t seem like a standard approach. Model-based filtering (C) is ambiguous here—could mean different things—so A seems like a safer bet given the focus on product preferences alongside user behavior.
B/C? Collaborative filtering (B) makes sense since it uses user similarity, but model-based filtering (C) can also be collaborative if it’s using latent factors. Option D isn’t really a standard term.
This one’s pretty clear because it’s about recommending products based on user similarity. Collaborative filtering (B) fits exactly here since it uses user-user relationships rather than just product features. Content-based (A) looks at product attributes, not similarities between users. Not sure about "model-based" as a separate category here, and I’ve never heard of “combinative filtering” (D). So B makes the most sense to me.