Machine Learning and Neural Networks AIE Practice Questions (Page 2)

Questions 6–10

1. 6foundation · easy

   Which pair of tasks correctly matches classification and regression?

   • AClassification: predicting the number of cars sold; Regression: identifying objects in an image.
   • BClassification: predicting whether an email is spam or not; Regression: predicting the price of a house.
   • CClassification: reducing the dimensionality of data; Regression: predicting a category label.
   • DClassification: predicting the temperature tomorrow; Regression: grouping customers by purchasing habits.
   Reveal Solution Discussion Select an answer first
2. 7application · medium

   A marketing team has a large dataset of customer purchase histories but no predefined customer segments. They want to identify natural groups of customers with similar buying behavior so the team can design targeted campaigns. Which machine learning approach should be used?

   • AReinforcement learning to optimize campaign selection
   • BSupervised classification to assign customers to known segments
   • CUnsupervised clustering to discover hidden customer segments
   • DSupervised regression to predict future purchase amounts
   Reveal Solution Discussion Select an answer first
4. 8expert · hard

   A manufacturing company wants to predict equipment failures before they occur. They have sensor data from machines, but only a few failures have been recorded historically, so the dataset is highly imbalanced. They also need the model to be interpretable so maintenance teams can understand why a failure is predicted. Which approach is most appropriate?

   • AUnsupervised anomaly detection to find unusual sensor readings
   • BReinforcement learning to optimize maintenance schedules
   • CA deep neural network with many hidden layers for high accuracy
   • DSupervised classification with a gradient-boosted tree model, using class weighting to handle imbalance
   Reveal Solution Discussion Select an answer first
5. 9application · easy

   A data analyst has a dataset of customer reviews and wants to automatically group them into topics (e.g., 'price', 'service', 'quality') without providing any predefined labels. Which type of machine learning is appropriate?

   • ASemi-supervised learning, using a small set of labels to guide the grouping
   • BSupervised learning, training a classifier on manually labeled reviews
   • CReinforcement learning, where the model learns to assign topics based on rewards
   • DUnsupervised learning, using topic modeling or clustering to discover hidden themes
   Reveal Solution Discussion Select an answer first
6. 10foundation · easy

   What is the primary purpose of an activation function in a neural network?

   • ATo reduce the number of parameters in the network.
   • BTo store the learned weights of the network.
   • CTo introduce non-linearity, allowing the network to learn complex patterns.
   • DTo normalize the input data to have zero mean and unit variance.
   Reveal Solution Discussion Select an answer first