Chapter 5 Logistic Regression

• Different from Naive Bayes which is a generative classifier, Logistic Regression is a discriminative classifier.

  • Generative model learns how a document of specific class is generated, or the distribution of documents over classes.

    • Given a test image, the system then asks whether it’s the cat model or the dog model that better fits (is less surprised by) the image, and chooses that as its label.

  • Discriminative model extracts features of different classes of documents that are the most helpful when classifying documents, instead of learning features on the whole.