An unsupervised self-learning system is used to define what the normal network looks like, and then uses this to detect deviations. Which category does this belong to?

• A. Supervised Learning
• B. Unsupervised Learning
• C. Classification
• D. Regression

Answer: (B)

Modeling normal behavior from unlabeled data and then flagging deviations is a hallmark of unsupervised learning. In this approach, you build a baseline of what typical network activity looks like without using labeled examples, then monitor new data to detect anomalies that don’t fit that baseline. This is exactly how anomaly detection works: learn the patterns of normal traffic and raise an alert when something unusual occurs.

Classification and regression, by contrast, rely on labeled data to map inputs to predefined categories or to predict continuous values, which isn’t what's described here.