# Formal definition for parameter setting in data mining context

While reading this material on decision trees, I came across the following statement:

The construction of decision tree classifiers does not require any domain knowledge or parameter setting, and therefore is appropriate for exploratory knowledge discovery.

I have couple of doubts related to this statement. First one is the definition of parameter setting in this context and another one is how being without parameter setting is apt for exploratory knowledge discovery?

First part about parameter setting regards to the fact that you don't need to define any parameters of the model, which is the case in per se linear regression which takes the form of ŷ =θ0+θ1x where $$θ_0,θ_1$$ are parameters. Decision trees create those n-dimensional boundaries for future classification (regression) based on certain criteria.