The availability of reliable learning systems is of strategic importance, as many tasks cannot be solved by classical programming techniques, because no mathematical model of the problem is available. So, for example, no one knows how to write a computer program that performs handwritten character recognition, though plenty of examples are available. It is, therefore, natural to ask if a computer could be trained to recognize the letter A from examples; after all, humans learn to read this way. Given the increasing quantity of data for analysis and the variety and complexity of data analysis problems being encountered in business, industry, and research, demanding the best solution every time is impractical. The ultimate dream, of course, is to have some intelligent agent that can preprocess data, apply the appropriate mathematical, statistical, and artificial intelligence techniques, and then provide a solution and an explanation. In the meantime, we must be content with the pieces of this automatic problem solver. The data miner’s purpose is to use the available tools to analyze data and provide a partial solution to a business problem.