Identify Supervised, Unsupervised, and Reinforcement Learning SNOWPRO-ADVANCED-DATA-SCIENTIST Practice Questions (Page 3)

Questions 11–15

1. 11application · medium

   A retail company wants to segment its customers into distinct groups based on purchasing behavior, demographics, and browsing history. There are no predefined customer segments or labels. The data science team needs to identify natural groupings to tailor marketing campaigns. Which learning approach should they use?

   • AUnsupervised learning with a clustering algorithm
   • BSupervised learning with a classification algorithm
   • CReinforcement learning with a reward function
   • DSupervised learning with a regression algorithm
   Reveal Solution Discussion Select an answer first
2. 12foundation · easy

   What is the primary goal of a reinforcement learning agent?

   • ATo discover hidden patterns in unlabeled data
   • BTo classify data into predefined categories
   • CTo predict a continuous value based on historical data
   • DTo learn optimal actions through trial and error to maximize cumulative reward
   Reveal Solution Discussion Select an answer first
4. 13expert · hard

   A recommendation system for an online video platform wants to learn which videos to suggest next. The system receives user interactions (clicks, watch time) and must decide the next video to recommend. The system is rewarded when users watch a video for a long time. Which learning approach is most appropriate?

   • ASemi-supervised learning with a small labeled set
   • BSupervised learning to predict watch time
   • CUnsupervised learning to cluster users
   • DReinforcement learning with watch time as the reward
   Reveal Solution Discussion Select an answer first
5. 14expert · hard

   A cybersecurity firm wants to develop an AI that learns to block network attacks. The AI receives network traffic data and must decide whether to block or allow each packet. The firm has historical logs of attacks, but the logs do not include the 'optimal' action for every situation. The AI will be rewarded for blocking attacks without blocking legitimate traffic. Which learning approach is most appropriate?

   • AReinforcement learning with a reward for correct blocking
   • BSupervised learning to classify attacks
   • CSemi-supervised learning with a small labeled set
   • DUnsupervised learning to cluster network traffic
   Reveal Solution Discussion Select an answer first
6. 15foundation · easy

   Which statement accurately describes unsupervised learning?

   • AIt requires labeled data to train a predictive model
   • BIt discovers hidden patterns or structures in unlabeled data
   • CIt uses an agent and reward signals to learn optimal actions
   • DIt maps inputs to known outputs using historical examples
   Reveal Solution Discussion Select an answer first