FANDOM


From Konstantinova (2014)[1]

"Singh et al. (2013) go even further and include coreference resolution as well. So they propose a single, joint graphical model that represents the various dependencies between the tasks (entity tagging, relation extraction, and coreference). Their joint modelling approach helps to avoid cascading errors. The joint model obtains 12 % error reduction on tagging over the isolated models."

Zitnik and Bajec (2015)[2]: "multiple natural language processing tasks, such as named entities recognition, relationships extraction and coreference resolution [...] In this paper we introduce a novel iterative and joint information extraction system that interconnects all the three tasks together using iterative feature functions which use the advantage of the intermediate extractions. [...] our model obtained a 15% error reduction on named entity recognition over individual models."

References Edit

  1. Konstantinova, Natalia. "Review of Relation Extraction Methods: What Is New Out There?." Analysis of Images, Social Networks and Texts. Springer International Publishing, 2014. 15-28.
  2. Žitnik, S., & Bajec, M. (2015). Iterative joint extraction of entities, relationships and coreferences from text sources. In 2015 IEEE 9th International Conference on Research Challenges in Information Science (RCIS) (pp. 412–422). inproceedings. http://doi.org/10.1109/RCIS.2015.7128902