[R] Model Selection and Data

[Melanie Vida]

Hi All,
   Are there times when random forests should not be used as a 
classification or regression model for determining variable importance. 
If so, then is it the properties of the training data that causes the 
issue? Are there other classification and regression models better 
suited for particular data sets? If so, what are the properties for the 
data set causing one to choose one model rather than another?
Thanks in advance,
-Melanie