Paper: Semi-Supervised Relation Extraction With Label Propagation

ACL ID N06-2007
Title Semi-Supervised Relation Extraction With Label Propagation
Venue Human Language Technologies
Session Short Paper
Year 2006
Authors

To overcome the problem of not hav- ing enough manually labeled relation in- stances for supervised relation extraction methods, in this paper we propose a label propagation (LP) based semi-supervised learning algorithm for relation extraction task to learn from both labeled and unla- beled data. Evaluation on the ACE corpus showed when only a few labeled examples are available, our LP based relation extrac- tion can achieve better performance than SVM and another bootstrapping method.