Selecting a Form of ML CT-AI Practice Questions (Page 6)

Questions 26–29

1. 26expert · hard

   An autonomous vehicle company is developing a system that must learn to navigate a city. The system can collect data from sensors and receives a reward for reaching its destination safely and a penalty for collisions. The company also has a large dataset of human driving recordings with labeled actions (e.g., turn left, brake). Which approach should the company take to train the navigation system?

   • AUse dimensionality reduction to simplify sensor data only
   • BUse supervised learning on the human driving dataset to predict actions
   • CUse unsupervised clustering to group similar driving scenarios
   • DUse reinforcement learning to learn a policy from the reward signals
   Reveal Solution Discussion Select an answer first
2. 27application · medium

   An e-commerce company wants to predict the exact dollar amount a customer will spend in the next month, based on their past purchase history. The target variable is a continuous monetary value. Which form of ML should be selected?

   • ASupervised regression
   • BUnsupervised clustering
   • CReinforcement learning
   • DSupervised classification
   Reveal Solution Discussion Select an answer first
4. 28application · medium

   A credit card company wants to build a model to approve or decline transactions in real time. They have a large dataset of past transactions with labels indicating whether each was fraudulent. The model must make a binary decision (approve/decline) quickly. Which form of ML is most appropriate?

   • AReinforcement learning to learn an approval policy
   • BUnsupervised clustering to group transactions
   • CSupervised regression to predict a fraud score
   • DSupervised classification to make the binary decision
   Reveal Solution Discussion Select an answer first
5. 29application · medium

   A manufacturing company has sensor data from machines, but no records of which machines have failed. The maintenance team wants to identify which machines are behaving abnormally so they can inspect them. Which form of ML should be used?

   • AUnsupervised anomaly detection to find unusual sensor patterns
   • BReinforcement learning to optimize maintenance schedules
   • CSupervised classification to predict machine failure
   • DSupervised regression to predict sensor values
   Reveal Solution Discussion Select an answer first