You have been tasked with creating a model that will recommend products based on what other customers have similarly purchased. Which algorithm is the best choice given this situation?

Product recommendation based on similarity to other customers’ purchase histories is a classic user-based collaborative filtering problem. A non-parametric K-Nearest Neighbor approach computes distance between customer vectors to surface nearest neighbors, yielding interpretable, fast recommendations without complex training.

B. K-means clusters customers but does not directly compute neighbor similarity for recommendations.
C. Neural networks can model recommendations but add unnecessary complexity and training cost.
D. Hyperpersonalization is a marketing strategy, not a specific algorithmic technique.

1. MIT OCW 6.0002, “Recommendation Systems,” Lecture 11 notes, pp. 3–6.

2. Ricci, Rokach & Shapira, Recommender Systems Handbook, 2nd ed., Ch. 4 “Neighborhood-based Methods,” pp. 107–109 (doi:10.1007/978-1-4899-7637-64).

3. Stanford CS246, Winter 2022, “Nearest Neighbors for Recommendations,” slides 15–20.