Bootstrapping via graph propagation

The Yarowsky algorithm is a simple self-training algorithm for bootstrapping learning from a small number of initial seed rules which has proven very effective in several natural language processing tasks. Bootstrapping a classifier from a small set of seed rules can be viewed as the propagation of labels between examples via features shared between them. This thesis introduces a novel variant of the Yarowsky algorithm based on this view. It is a bootstrapping learning method which uses a graph propagation algorithm with a well-defined objective function. The experimental results show that our proposed bootstrapping algorithm achieves state of the art performance or better on several different natural language data sets.