The Tradeoffs Between Open and Traditional Relation Extraction

@inproceedings{Banko2008TheTB,
  title={The Tradeoffs Between Open and Traditional Relation Extraction},
  author={Michele Banko and Oren Etzioni},
  booktitle={ACL},
  year={2008}
}
Traditional Information Extraction (IE) takes a relation name and hand-tagged examples of that relation as input. Open IE is a relationindependent extraction paradigm that is tailored to massive and heterogeneous corpora such as the Web. An Open IE system extracts a diverse set of relational tuples from text without any relation-specific input. How is Open IE possible? We analyze a sample of English sentences to demonstrate that numerous relationships are expressed using a compact set of… CONTINUE READING

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

Key Quantitative Results

  • We present H-CRF, an ensemble-based extractor that learns to combine the output of the lexicalized and unlexicalized RE systems and achieves a 10% relative increase in precision with comparable recall over traditional RE.

Citations

Publications citing this paper.
SHOWING 1-10 OF 293 CITATIONS

Open Question Answering

VIEW 11 EXCERPTS
CITES METHODS & BACKGROUND
HIGHLY INFLUENCED

Extracting Meta Statements from the Blogosphere

VIEW 5 EXCERPTS
CITES METHODS & BACKGROUND
HIGHLY INFLUENCED

RDR-based open IE for the web document

VIEW 19 EXCERPTS
CITES METHODS & BACKGROUND
HIGHLY INFLUENCED

Machine Learning for Text

  • Springer International Publishing
  • 2018
VIEW 5 EXCERPTS
HIGHLY INFLUENCED

A hybrid ontology-based information extraction system

VIEW 13 EXCERPTS
CITES METHODS & BACKGROUND
HIGHLY INFLUENCED

Co-occurrence rate networks: towards separate training for undirected graphical models

VIEW 10 EXCERPTS
CITES RESULTS, BACKGROUND & METHODS
HIGHLY INFLUENCED

Open information extraction based on lexical semantics

  • Journal of the Brazilian Computer Society
  • 2015
VIEW 10 EXCERPTS
CITES BACKGROUND
HIGHLY INFLUENCED

FILTER CITATIONS BY YEAR

2008
2019

CITATION STATISTICS

  • 54 Highly Influenced Citations

  • Averaged 10 Citations per year from 2017 through 2019