Are decision forests and random forests the same thing?

Question

Based on an answer on Cross Validated I've been looking into implementing a random forest in .NET/C# to classify documents of text.

Looking around the web to see if there are existing implementations, I came across an algorithm for a decision forest on Alglib.

The thing is, I can't seem to find anything specific to "decision forest" anywhere (it's ambiguous even here), it's typically mixed in with random forest.

That said, are the two the same, just being referenced differently, or are there inherent differences between the two?

Answer 1

At the Alglib page you cited, it says,

"The RDF [Random decision forest] algorithm is a modification of the original Random Forest algorithm designed by Leo Breiman and Adele Cutler."

A question at rapid-i.com refers to

Ho TK (1998) The Random Subspace Method for Constructing Decision Forests. IEEE Trans Pattern Anal Mach Intel 20(8) 832-844 [Abstract]

which might be yet another thing.

At Breiman's web page at Berkeley, it says,

"Random Forests(tm) is a trademark of Leo Breiman and Adele Cutler and is licensed exclusively to Salford Systems for the commercial release of the software. Our trademarks also include RF(tm), RandomForests(tm), RandomForest(tm) and Random Forest(tm)."

So I conclude that there are subtle differences, but mostly it's a trademark issue regarding the name "random forest".