Verification goal of KDD? Cases where it apply I'm having difficulties in finding examples of a Verification Data-mining/KDD approach. Let me explain what I mean by that.
In the notorious article From Data Mining to KDD by Fayyad et al the following definition is mentioned: 
"The knowledge discovery goals are defined by the intended use of the system. We can distinguish two types of goals: (1) verification and (2) discovery."
After, the Discovery goal is then divided in prediction and description categories . 
But, I'm using as my main source the book Introduction to Data Mining by Tan et al, and there he doesn't mention the verification goal of Data-mining. In addition, the web doesn't seems to help either. Could you please give me a hand on finding cases where the goal of the process is verification?

Optional:
I have some hypothesis. In this article David J. Hand says:
"It is probably no exaggeration to say to say that most statisticians are concerned with primary data analysis. That is, the data are collected with a particular question or set of question in mind.".
Thus, could this scope be one example where the verification goal could be applied? 
 A: I think you've got it.
The authors seem to divide 'Knowledge Discovery from Databases (KDD)' into verification, where the purpose of the work is to test a specific, pre-specified hypothesis, and discovery, where the goal is to find hitherto-unknown relationships in the data. These are sometimes called confirmatory data analysis and exploratory data analysis, respectively. 
Here's a fairly memorable example. In 1991, Andrew Bell, David Brown, and Nicholas Terrett developed compound UK-92-480, a phosphodiesterase inhibitor. Based on what was known about the heart, they hypothesised that this compound could dilate veins and arteries while also preventing platelets from clumping together, all of which might help treat certain cardiac conditions. Several clinical trials studied how varying doses of the drug affected subjects' blood pressure. Dose and blood pressure values, along with other information (including side effects) were recorded. 
The "verification" part of their analysis looked for a relationship between the drug dose and blood pressure. They found that the drug was only weakly effective at controlling blood pressure, especially at doses patients could tolerate. However, they noticed an interesting pattern of side effects in the male subjects. Based on the results of this discovery, Pfizer repurposed the drug. After a series of fantastically successful clinical trials, it was (re)named 'Viagra' and made them nearly two billion dollars in 2008.
This story highlights an important point about the distinction between the two regimes. You can consider lots of hypotheses when exploring, but the things you "discover" need to be validated using different data. 
