We will first present a common problem, which we will use as a challenge to test six different classification algorithms. This common problem is referred to as the classifier challenge in this chapter. Using all the six classifiers to solve the same problem will help us in two ways:

• All the input variables need to be processed and assembled as a complex data structure, called a feature vector. Using the same feature vector helps us avoid repeating data preparation for all six algorithms. 
• We can compare the performance of various algorithms as we are using the same feature vector for input. 

The classifiers challenge is about predicting the likelihood of a person making a purchase. In the retail industry, one of the things that can help maximize sales is better understanding the behavior of the customers. This can be done by analyzing the patterns found in historical data. Let's state the problem, first.