Identify Common Machine Learning Techniques AI-900 Practice Questions (Page 3)

Questions 11–15

1. 11application · medium

   A logistics company wants to predict the delivery time (in minutes) for each package based on distance, traffic conditions, and weather. They also want to identify packages that are likely to be delayed significantly. Which approach should they use?

   • AUse classification to predict 'on-time' or 'delayed', then use regression to find the delay amount.
   • BUse clustering to group packages by route, then use classification to label each package as 'delayed'.
   • CUse a transformer to generate a text report of delivery times.
   • DUse regression to predict delivery time, then use anomaly detection to flag unusually long predicted times.
   Reveal Solution Discussion Select an answer first
2. 12application · medium

   A news organization wants to automatically summarize long articles and answer questions based on their content. The system must understand context and relationships between words across the entire text. Which architecture is best suited for this task?

   • AA Transformer model with self-attention mechanisms.
   • BA regression model to predict article length.
   • CA clustering algorithm to group similar articles.
   • DA simple feedforward neural network with one hidden layer.
   Reveal Solution Discussion Select an answer first
4. 13application · medium

   A company wants to build a system that can recognize objects in real-time video streams, such as identifying cars, pedestrians, and traffic signs. The system must handle varying lighting conditions and occlusions. Which approach is most appropriate?

   • AUse a clustering algorithm to group similar video frames.
   • BUse a transformer model to process the video as a sequence of text descriptions.
   • CUse a deep neural network with multiple layers trained on labeled images.
   • DUse a regression model to predict the number of objects in each frame.
   Reveal Solution Discussion Select an answer first
5. 14application · medium

   A social media platform wants to automatically detect hate speech and offensive content in user posts. The system must understand context and sarcasm. Which architecture is best suited?

   • AA regression model to predict a toxicity score.
   • BA Transformer model fine-tuned on labeled posts.
   • CA simple keyword matching system.
   • DA clustering algorithm to group similar posts.
   Reveal Solution Discussion Select an answer first
6. 15expert · hard

   A manufacturing company wants to detect defective products on an assembly line using images. They have a large dataset of images of good products but very few images of defective products. Which approach is most appropriate?

   • AUse a clustering algorithm to group images of good products and flag images that are outliers as defective.
   • BUse a regression model to predict a defect score for each image.
   • CUse a classification model trained on the available images, including the few defective ones.
   • DUse a deep learning model with data augmentation to increase the defective samples.
   Reveal Solution Discussion Select an answer first