9 Laws of Data Mining

Theses laws were established by Thomas Khabaza, a pioneer of data mining.

1) Business Goals
business objectives are at the origin of every data-mining solution
You should focus on business goals throughout the data-mining process.

2) Business Knowledge
business knowledge is central to every step of the data-mining process
Data mining is only valuable if you have the business understanding to give context to your data.

3) Data Preparation
data preparation is more than half of every data-mining process
Data usually isn't collected specifically for data mining.

4) Right Model (no free lunch)
the right model for a given application can only be discovered by experiment
Models are selected through trial and error rather than by studying theory.

5) Pattern
there are always patterns
Data always has something to say, whether it's what you wanted to hear or not.

6) Amplification (insight)
data mining amplifies perception in the business domain
Data mining finds information that wouldn't appear in ordinary reports.

7) Prediction
prediction increases information locally by generalization
Data mining uses what we know to predict what we don't know.

8) Value
the value of data-mining results is not determined by the accuracy or stability of predictive models
Priority is not given to theoretical properties, but rather to business applications.

9) Change
all patterns are subject to change
Models you create today might be useless tomorrow.