Design an item category prediction system

Q: Design an item category prediction system

This question evaluates a candidate's ability to design end-to-end machine learning systems for multi-class or hierarchical item categorization, encompassing competencies in feature engineering, multimodal modeling, labeling and data quality strategies, evaluation metrics, serving architecture, scalability, and operational monitoring.

Q: How do I approach ML System Design interview questions?

ML System Design questions require understanding of core concepts and practice. PracHub provides solutions with explanations to help you master ml system design interviews.

Question

Loading...

Design an end-to-end ML system that predicts an item’s category/type (multi-class or hierarchical classification), e.g., assigning an e-commerce listing to a taxonomy like Electronics > Headphones.

What to cover

Inputs available at prediction time (title, description, images, attributes, seller signals, etc.)
Training data and labeling strategy (including noisy labels)
Model choices (text-only vs multimodal; flat vs hierarchical)
Offline evaluation metrics and online success metrics
Serving architecture and latency/throughput targets
Handling cold-start items, new categories, and class imbalance
Monitoring, drift detection, and retraining/feedback loops
Human-in-the-loop / moderation workflow for low-confidence cases

Design an item category prediction system

Quick Overview

What to cover

Solution

Comments (0)