Paper: Coreference for Learning to Extract Relations: Yes Virginia Coreference Matters

ACL ID P11-2050
Title Coreference for Learning to Extract Relations: Yes Virginia Coreference Matters
Venue Annual Meeting of the Association of Computational Linguistics
Session Main Conference
Year 2011
Authors

As an alternative to requiring substantial su- pervised relation training data, many have ex- plored bootstrapping relation extraction from a few seed examples. Most techniques assume that the examples are based on easily spotted anchors, e.g., names or dates. Sentences in a corpus which contain the anchors are then used to induce alternative ways of expressing the relation. We explore whether coreference can improve the learning process. That is, if the algorithm considered examples such as his sister, would accuracy be improved? With co- reference, we see on average a 2-fold increase in F-Score. Despite using potentially errorful machine coreference, we see significant in- crease in recall on all relations. Precision in- creases in four cases and decreases in six.