Explain Core AI Principles and Terminology CY0-001 Practice Questions (Page 6)

Questions 26–30

1. 26application · medium

   A security team is training a machine learning model to detect phishing URLs. They split their labeled dataset into a training set and a test set. After training, they evaluate the model on the test set and achieve 99% accuracy. What is the most important reason for using a separate test set?

   • ATo assess how well the model generalizes to new, unseen data
   • BTo eliminate the need for feature engineering
   • CTo increase the amount of data available for training
   • DTo reduce the training time of the model
   Reveal Solution Discussion Select an answer first
2. 27foundation · easy

   What is the basic structure of a neural network?

   • AA database of known patterns that are matched against new input.
   • BA single layer of interconnected nodes that process data in parallel.
   • CA collection of decision trees that vote on the final output.
   • DA set of layers of interconnected nodes (neurons) that transform input data through weighted connections.
   Reveal Solution Discussion Select an answer first
4. 28foundation · easy

   What is a 'label' in supervised machine learning?

   • AA type of activation function.
   • BThe known correct output for a given input, used to train the model.
   • CA unique identifier for the model version.
   • DA security classification assigned to a network segment.
   Reveal Solution Discussion Select an answer first
5. 29application · medium

   A security vendor claims their new tool can analyze phishing emails by understanding the intent and sentiment of the message text, not just matching keywords. Which AI capability is the tool primarily using?

   • AAnomaly detection
   • BNatural language processing (NLP)
   • CReinforcement learning
   • DComputer vision
   Reveal Solution Discussion Select an answer first
6. 30foundation · easy

   Why is ethical consideration important when deploying AI in cybersecurity?

   • ABecause AI models are always correct and do not need human oversight.
   • BBecause AI decisions can have significant impacts on individuals' privacy and civil liberties.
   • CBecause AI systems are not subject to any regulations.
   • DBecause AI models cannot be biased if they are trained on large datasets.
   Reveal Solution Discussion Select an answer first