Classifying email as spam, labeling web pages based on their context, voice recognition are examples of which of the following?

Explanation

Spam classification, topic/context labeling of web pages, and voice/speech recognition are all tasks where a model is trained on examples that already carry the correct output label (spam/not-spam, the topic label, the spoken word), and the model learns a mapping from input to known output. This is the defining characteristic of supervised learning, unlike reinforcement learning (reward-based, no labelled examples), unsupervised learning (no labels at all) or a shortest path algorithm (not a learning method). Hence option (3) is correct.