You work for a company that is developing a new video streaming platform. You have been asked to create a recommendation system that will suggest the next video for a user to watch. After a review…

Question

Options

• ALaunch the product without machine learning. Present videos to users alphabetically, and start
• BLaunch the product without machine learning. Use simple heuristics based on content metadata
• CLaunch the product with machine learning. Use a publicly available dataset such as MovieLens to
• DLaunch the product with machine learning. Generate embeddings for each video by training an

• A
  3% (1)
• B
  79% (23)
• C
  3% (1)
• D
  14% (4)

Explanation

Without any historical user event data (views, clicks, watch history), there is no signal to train a personalized ML recommendation model. Attempting ML without training data would be premature and unreliable. The pragmatic first-version approach is to use simple heuristics based on available content metadata (content type, release date, country) to surface relevant videos while the platform collects user interaction data. Option A (alphabetical) provides no personalization value. Option C (MovieLens) uses a different domain's data with different user behavior patterns. Option D requires user interaction data that doesn't yet exist.

No community discussion yet for this question.