Methodology of Application
As mentioned in Chapter 1 above, the goal of the information-theoretic methodology, presented in this book, is to automate the entire process of knowledge discovery. Though this is a first attempt to create a unified framework for most KDD stages, we are far from providing a “turn-key” solution to such a complex problem as building models of knowledge from data. The information-theoretic techniques, presented here, and their results (in the form of networks, rules, predictions, etc.) are just a part of a larger process, which is aimed at identifying and solving business problems.
KeywordsDimensionality Reduction Target Attribute Input Attribute Nominal Attribute Data Mining Algorithm
Unable to display preview. Download preview PDF.