In 1992, Eric Brill introduced the formalism of the transformation-based learning algorithm in its current formulation, but probably a more well-known reference to the technique is Eric Brill's article from 1995. The central idea behind transformation based learning (henceforth TBL) is to start with some simple solution to the problem, and apply transformations - at each step the transformation which results in the largest benefit is selected and applied to the problem. The algorithm stops when the selected transformation does not modify the data in enough places, or there are no more transformations to be selected.
Boosting is a general method of producing a very accurate prediction rule by combining rough and moderately inaccurate "rules of thumb." Most recent work has been on the "AdaBoost" boosting algorithm and its extensions.
|ID3 & C4.5|