Learning 5000 Relational Extractors

@inproceedings{Hoffmann2010Learning5R,
  title={Learning 5000 Relational Extractors},
  author={Raphael Hoffmann and Congle Zhang and Daniel S. Weld},
  booktitle={ACL},
  year={2010}
}
Many researchers are trying to use information extraction (IE) to create large-scale knowledge bases from natural language text on the Web. However, the primary approach (supervised learning of relation-specific extractors) requires manually-labeled training data for each relation and doesn't scale to the thousands of relations encoded in Web text. This paper presents LUCHS, a self-supervised, relation-specific IE system which learns 5025 relations --- more than an order of magnitude greater… CONTINUE READING

Figures, Tables, Results, and Topics from this paper.

Key Quantitative Results

  • This paper presents LUCHS, a self-supervised, relation-specific IE system which learns 5025 relations --- more than an order of magnitude greater than any previous approach --- with an average F1 score of 61%.
  • This paper presents LUCHS, a self-supervised, relation-specific IE system which learns 5025 relations — more than an order of magnitude greater than any previous approach — with an average F1 score of 61%.
  • We evaluate the overall end-to-end perfor- mance of LUCHS, showing an F1 score of 61% when extracting relations from randomly selected Wikipedia pages.
  • We show an overall performance of 61% F1 score, and present experiments evaluating LUCHS’s individual components.

Citations

Publications citing this paper.
SHOWING 1-10 OF 113 CITATIONS

Information Access Evaluation. Multilinguality, Multimodality, and Interaction

  • Lecture Notes in Computer Science
  • 2014
VIEW 10 EXCERPTS
CITES METHODS, BACKGROUND & RESULTS
HIGHLY INFLUENCED

The Semantic Web – ISWC 2012

  • Lecture Notes in Computer Science
  • 2012
VIEW 10 EXCERPTS
CITES BACKGROUND & METHODS
HIGHLY INFLUENCED

Bootstrapping for Numerical Open

IE Swarnadeep Saha
  • 2017
VIEW 3 EXCERPTS
CITES BACKGROUND & RESULTS
HIGHLY INFLUENCED

Bootstrapping for Numerical Open IE

VIEW 3 EXCERPTS
CITES BACKGROUND & RESULTS
HIGHLY INFLUENCED

Omni-word Feature and Soft Constraint for Chinese Relation Extraction

VIEW 4 EXCERPTS
CITES METHODS & BACKGROUND
HIGHLY INFLUENCED

FILTER CITATIONS BY YEAR

2010
2019

CITATION STATISTICS

  • 11 Highly Influenced Citations

  • Averaged 5 Citations per year from 2017 through 2019

References

Publications referenced by this paper.
SHOWING 1-10 OF 35 REFERENCES

Semi-supervised acquisition of semantic classes – from the web and for the web

Ye-Yi Wang, Raphael Hoffmann, Xiao Li, Alex Acero.
  • International Conference on Information and Knowledge Management (CIKM2009), pages 37–46.
  • 2009
VIEW 1 EXCERPT