The Fundamentals of Classification

While regression focuses on creating a model that best fits our data to predict the future, classification is all about creating a model that separates our data into separate classes.

Assuming that you have some data belonging to separate classes, classification helps you predict the class a new data point belongs to. A classifier is a model that determines the label value belonging to any data point in the domain. Suppose you have a set of points, P = {p1, p2, p3, ..., pm}, and another set of points, Q = {q1, q2, q3, ..., qn}. You treat these points as members of different classes. For simplicity, we could imagine that P contains credit-worthy individuals, and Q contains individuals that are risky in terms of their credit repayment tendencies.

You can divide the state space so that all points in P are on one cluster of the state space, and then disjoint from the state space cluster containing all points in Q. Once you find these bounded spaces, called clusters...