We propose a general approach to modeling semi-supervised learning constraints on unlabeled data. Both traditional supervised classification tasks and many natural semi-supervised learning heuristics can be approximated by specifying the desired outcome of walks through a graph of classifiers. We demonstrate the modeling capability of this approach in the task of relation extraction, and experimental results show that the modeled constraints achieve better performance as expected.
Paper
Full text
Using Graphs of Classifiers to Impose Constraints on Semi-supervised Relation Extraction
Semantic Scholar · Computer Science · 2016
Abstract
We propose a general approach to modeling semi-supervised learning constraints on unlabeled data. Both traditional supervised classification tasks and many natural semi-supervised learning heuristics can be approximated by specifying the desired outcome of walks through a graph of classifiers. We demonstrate the modeling capability of this approach in the task of relation extraction, and experimental results show that the modeled constraints achieve better performance as expected.