II STREAM MINING

Classification is one of the most widely used data mining techniques. In very general terms, given a list of groups (often called classes), classification seeks to predict which group a new instance may belong to. The outcome of classification is typically either to identify a single group or to produce a probability distribution of the likelihood of membership for each group. It is the most important case of supervised learning, meaning that the labels available to the learning algorithm can be used to guide it. A spam filter is a good example, where we want to predict whether new emails are considered spam or not. Twitter sentiment analysis is another example, where we want to predict whether the sentiment of a new incoming tweet is positive or negative.