6.4 Explain AI (generative and Predictive) and Machine Learning in Network Operations 200-301 Practice Questions (Page 1)

Questions 1–5

1. 1foundation · easy

   Which machine learning approach is most appropriate for discovering groups of similar devices in a network without predefined categories?

   • ASupervised learning
   • BPredictive analytics
   • CReinforcement learning
   • DUnsupervised learning
   Reveal Solution Discussion Select an answer first
2. 2application · medium

   A network team wants to use predictive AI to forecast bandwidth usage on a WAN link. They have collected two years of traffic data, but the data contains many gaps and inconsistent timestamps due to monitoring tool changes. What is the most important first step before training the model?

   • AUse a generative AI model to create synthetic data to replace the gaps.
   • BAbandon the project because AI cannot handle imperfect data.
   • CTrain the model immediately to see what patterns it finds.
   • DClean and preprocess the data to fill gaps and standardize timestamps.
   Reveal Solution Discussion Select an answer first
4. 3foundation · easy

   A network team has a labeled dataset of network traffic where each sample is marked as 'normal' or 'attack'. They train a model to classify new traffic. Which type of machine learning is this?

   • AUnsupervised learning
   • BReinforcement learning
   • CSupervised learning
   • DGenerative learning
   Reveal Solution Discussion Select an answer first
5. 4expert · hard

   A network team wants to implement an AI system that learns to reroute traffic around congested links. They have a network simulator that can provide feedback on the impact of routing decisions. However, they are concerned about the time it takes to train the model. Which approach is most suitable?

   • AGenerative AI to create routing tables
   • BUnsupervised learning to cluster traffic patterns
   • CReinforcement learning with a simulated environment for training
   • DSupervised learning with historical routing decisions
   Reveal Solution Discussion Select an answer first
6. 5application · medium

   A network engineer needs to quickly generate a standard ACL configuration for a new branch office. The engineer has a set of previously approved ACL templates and wants to use a generative AI tool to produce a draft. What is the most important step before applying the AI-generated ACL to production?

   • AImmediately apply the ACL to the production router to save time
   • BReview the generated ACL against the approved templates and test it in a lab environment
   • CUse the AI-generated ACL only if it matches the exact syntax of a previous ACL
   • DAsk the AI to generate multiple versions and pick the longest one
   Reveal Solution Discussion Select an answer first