Free Google Professional-Machine-Learning-Engineer Actual Exam Questions - Question 3 Discussion

C imo. Collaborative filtering is the classic choice for recommendations based on user similarity and purchase behavior, even if ratings aren’t explicit. It naturally captures patterns in user interactions.

C/D? Collaborative filtering fits the user similarity part, but if we just have purchase amounts or frequencies, regression might help predict preferences too. Depends on what data is really available.

B imo. The question says "based on purchase behavior and similarity," but it doesn’t mention ratings or large user data needed for collaborative filtering. Knowledge-based filtering can work well when you have specific product features and user preferences, especially if the purchase data is sparse or new users pop up frequently. It’s more straightforward to implement without needing tons of historical user data. So, it might be safer than jumping straight to collaborative-based filtering, which usually needs richer data to perform well.

Yeah, collaborative filtering (C) fits since it uses patterns from user behavior and similarities to recommend items. Classification (A) and regression (D) don’t really focus on the recommendation aspect. Knowledge-based filtering (B) usually needs explicit user preferences, which we don’t have here. So, C makes the most sense given the info.

C. Since the question highlights similarity with other users and purchase behavior, collaborative filtering is the most straightforward approach for recommendations, unlike classification or regression which aren’t designed for this task.

It’s C because collaborative filtering directly tackles recommendations using user similarity and past purchases, unlike A or D which are more about predicting specific labels or values, not recommending products.

This is a tricky one. While C fits with the idea of user similarity, B could also be relevant if the purchase data is limited or sparse since knowledge-based filtering relies more on explicit user preferences or product attributes. The question doesn’t clarify the amount of historical data, so B might be safer if we can’t assume enough user interactions for collaborative filtering. Still, since it explicitly mentions similarity with other users, C feels more aligned overall. But I wouldn’t completely rule out B without knowing the data size. So I’m sticking with C for now.

C imo, since it’s all about user similarity and behavior patterns, classic collaborative filtering.

Option C sounds right, but another angle is that classification or regression (A or D) don’t capture user similarity well, so they can be ruled out quickly. Collaborative filtering is built for these recommendation scenarios.

The question focuses on recommending products based on user similarity and purchase behavior, which fits the collaborative filtering approach best. So, I’d go with C.

C, since it matches the requirement of using user similarity and purchase behavior directly.

Maybe B, since knowledge-based filtering can also use user preferences without needing lots of data.

C imo, because collaborative filtering directly uses user behavior and similarity, which is exactly what the question asks for. Classification or regression don’t really handle recommendations based on user-user similarity.

C