Common AI Technologies in Daily Life AIE Practice Questions (Page 4)

Questions 16–20

1. 16expert · hard

   A streaming service wants to improve recommendations for a new user who has just signed up and has not yet watched anything. The service has a large catalog with rich metadata (genre, director, actors) and also has aggregated viewing data from millions of existing users. The team must choose a strategy that works immediately for the new user and also improves as the user watches more content. Which strategy best satisfies both constraints?

   • AStart with content-based filtering using the user's selected preferences and catalog metadata, then blend in collaborative filtering as viewing history accumulates
   • BUse only collaborative filtering on the aggregated viewing data, assuming the new user will behave like the average user
   • CShow the same globally popular titles to all new users until they have watched at least 10 items
   • DUse only content-based filtering based on catalog metadata, ignoring all viewing history from other users
   Reveal Solution Discussion Select an answer first
2. 17foundation · easy

   How is AI used to control non-player characters (NPCs) in video games?

   • ABy requiring a human to control each NPC
   • BBy using random movement patterns
   • CBy pre-recording all possible NPC movements
   • DBy enabling NPCs to make decisions and react to player actions
   Reveal Solution Discussion Select an answer first
4. 18expert · hard

   A navigation app is being redesigned. The team wants to provide real-time traffic predictions and also support an autonomous vehicle mode that can make driving decisions. The team has limited computational resources on the vehicle. Which approach best balances the need for real-time predictions and autonomous decision-making?

   • AUse cloud-based models for both traffic prediction and driving decisions to ensure accuracy
   • BUse collaborative filtering to predict traffic based on other users' routes and on-device models for driving decisions
   • CUse cloud-based machine learning for traffic prediction and on-device lightweight models for immediate driving decisions
   • DUse on-device models for both traffic prediction and driving decisions to avoid network latency
   Reveal Solution Discussion Select an answer first
5. 19foundation · easy

   How do AI chatbots enhance customer service on e-commerce websites?

   • ABy providing instant responses to common customer queries
   • BBy replacing all human customer service agents
   • CBy requiring customers to fill out long forms
   • DBy randomly selecting answers from a database
   Reveal Solution Discussion Select an answer first
6. 20application · medium

   An email client wants to reduce the time users spend composing replies. The client has access to the user's past email responses and the content of the incoming message. Which AI feature would most directly help the user respond quickly?

   • ASpam filtering that moves promotional emails to a separate folder
   • BSmart replies generated by natural language models that suggest short, contextually appropriate responses
   • CCollaborative filtering that recommends emails similar users replied to
   • DObject detection to identify attachments in the incoming email
   Reveal Solution Discussion Select an answer first