Unsupervised Learning CT-AI Practice Questions (Page 6)

Questions 26–30

1. 26application · medium

   A manufacturing company monitors sensor data from machines to detect equipment failures. They have historical data from normal operation but very few examples of failures. They want to flag unusual sensor readings that may indicate a developing fault. Which approach is most appropriate?

   • APCA to reduce sensor data dimensions
   • BAnomaly detection on the sensor data
   • CSupervised classification using the few failure examples
   • DAssociation rule mining on sensor readings
   Reveal Solution Discussion Select an answer first
2. 27foundation · easy

   What is the goal of association rule learning in unsupervised learning?

   • ATo reduce the dimensionality of the feature space.
   • BTo group data points into clusters based on similarity.
   • CTo predict the exact value of a continuous variable.
   • DTo discover interesting relationships or patterns between items in a dataset.
   Reveal Solution Discussion Select an answer first
4. 28expert · hard

   A data scientist is evaluating two unsupervised learning tasks on the same unlabeled dataset: (1) grouping similar records and (2) finding items that frequently co-occur. Which pair of techniques is appropriate for these tasks, respectively?

   • AAnomaly detection and k-means clustering
   • BPCA and t-SNE
   • CHierarchical clustering and PCA
   • Dk-means clustering and association rule mining
   Reveal Solution Discussion Select an answer first
5. 29expert · hard

   A data analyst is using k-means clustering on a dataset and notices that the resulting clusters are highly sensitive to the initial random seed. They run the algorithm multiple times and get different cluster assignments. Which approach is most appropriate to address this issue?

   • AIncrease the number of clusters k to reduce sensitivity.
   • BRun k-means multiple times with different seeds and select the result with the lowest within-cluster sum of squares (WCSS).
   • CSwitch to supervised classification to avoid randomness.
   • DUse PCA to reduce dimensions before clustering.
   Reveal Solution Discussion Select an answer first
6. 30foundation · easy

   Which of the following is a real-world application of unsupervised learning?

   • ADiagnosing a disease from labeled medical images.
   • BCustomer segmentation based on purchasing behavior.
   • CEstimating the price of a used car from its features.
   • DPredicting whether an email is spam or not.
   Reveal Solution Discussion Select an answer first