Evaluate Product-Ranking Algorithm with Precision and Recall Metrics

Q: Evaluate Product-Ranking Algorithm with Precision and Recall Metrics

This question evaluates competency in assessing product‑ranking models with emphasis on ranking metrics, precision and recall interpretations, experiment metric design, and the ability to map model outputs to user‑level funnels and business KPIs.

Q: How do I approach Machine Learning interview questions?

Machine Learning questions require understanding of core concepts and practice. PracHub provides solutions with explanations to help you master machine learning interviews.

Question

Scenario

Instagram Shopping wants to improve its product‑ranking algorithm for the shopping feed. The goal is to select and order products for each user to maximize value (e.g., conversions/revenue) while maintaining a good user experience.

Task

Propose offline metrics to evaluate a ranking model for a shopping feed and explain why they fit this use case.
Propose online (experiment) metrics, including how you would structure user‑level funnels.
Explain how precision and recall translate to ranking in this setting (including @K variants).
Explain how you would reconcile offline metrics with on‑site business KPIs.

Hints (for scope)

Consider NDCG and MAP for offline ranking evaluation.
Consider click‑through, conversion lift, and funnels for online evaluation.

Evaluate Product-Ranking Algorithm with Precision and Recall Metrics

Scenario

Task

Hints (for scope)

Solution

Comments (0)

Evaluate Product-Ranking Algorithm with Precision and Recall Metrics

Overview

Scenario

Task

Hints (for scope)

Solution

Comments (0)