Katkat
Back

Design a Family-Friendly Listing Classifier

AirbnbMachine Learning Engineer · Onsite
MediumMachine Learning EngineerOnsiteML System Design

Last updated

This question evaluates proficiency in machine learning system design, including problem formulation tied to product objectives, labeling strategy, feature engineering, model selection, evaluation metrics, robustness to adversarial labels, fairness considerations, and cold-start handling.

Design a machine learning system for a short-term rental marketplace that classifies whether a property listing is suitable for families. Users should be able to filter search results to show family-friendly homes.

Address the following:

  1. Define the prediction problem and product objective.
  2. Construct labels for family-friendly listings.
  3. Propose useful features.
  4. Choose baseline and advanced models.
  5. Define offline and online evaluation metrics.
  6. Handle hosts who intentionally mislabel their properties.
  7. Avoid unfair or discriminatory filtering.
  8. Handle cold-start listings with little or no historical data.