Introduction to Directed Data Mining
Directed data mining is the most common type of data mining. “Directed” means that the historical data used for modeling contains examples of the target values, so the data mining techniques have examples to learn from. Directed data mining also makes the assumption that the patterns in historical data are applicable in the future.
Another type of data mining is undirected data mining, which uses sophisticated techniques to find patterns in the data without the guidance of a target. Without a target, the algorithm cannot determine if the results are good or bad; as a consequence, undirected data mining requires additional human judgment to assess the results. Association rules are one example of undirected data mining. Other undirected techniques are typically more specialized, so this chapter and the next two focus on directed techniques.