Not everyone is well versed in the area of data mining, so this chapter starts straight away with what it is and what it is generally used for. So, without further ado… Data mining is the process of applying algorithms to data sets with the goal of exposing patterns in the data that would not otherwise be noticed. The reason such patterns would not otherwise be noticed owes to the complexity and volume of information within which the patterns are embedded. Another, less academic way to look at data mining is as a technology that can be used to answer questions like the following:
When customers visit our corporate web site, what paths are they most likely to take when navigating through the site?
When a $10 credit card transaction is processed at a gas station immediately followed by a $600 purchase on the same account from an electronics store in a different zip code, should a red flag be raised?
For optimal sales revenue generation in a grocery store, which products should be placed in close proximity to one another?
To address these types of questions, and many others, turn to data mining technology. What is coming up in this chapter on data mining will teach you more about the types of questions that can be asked and answered in the real world by creating and using data mining applications. You look at each algorithm in detail, and then learn about mining models, which can be built on top of cubes (OLAP Mining Models) or on top of raw relational data (relational mining models).