CMPSC 165A Lecture Notes - Lecture 6: Posterior Probability, Bayesian Network, Generative Model
Document Summary
A description of an instance, , where is the instance space. The category of :() , where () is a categorization function whose domain is and whose range is. Want to know how to build categorization functions (classifiers) If all hypotheses are a priori equally likely, we only need to consider the (| ) term: 2a argmax. Learning and classification methods based on probability theory. Build generative model that approximates how data is produced. Categorization produces a posterior probability distribution over the possible categories given a description of an item. Task: classify a new instance based on a tuple of attribute values = (,*, ,, into one of the classes e . Can be estimated from the frequency of classes in the training examples. Could only be estimated if a very large number of training examples was available. Assume that the probability of observing the conjunction of attributes is equal to the product of the individual probabilities.
