In the past two sessions, we talked about grammatical and rule-based concept extraction. In both cases, the task involved identifying and labeling instances of entities that fit a certain pattern. This session, we look at rule-based classification, which is more complex than concept extraction. Classification is a task which involves looking at entities, taking stock of their attributes, and then classifying them into categories. This is a complex topic, so we will be breaking it into two sessions; the first will be largely theoretical and the second will get into some of the technical details of how machine learning, the method we use to automate classification tasks, works.