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.

[跟我联系|返回]

2002 May,22