Relation Extraction from Large-scale Corpus

Date: May 2, 2014

Speaker: Yuhao Zhang, BMIR, Stanford University

Abstract

Pattern-based:

• Automatic acquisition of hyponyms from large text corpora (http://delivery.acm.org/10.1145/1000000/992154/p539-hearst.pdf)

Supervised learning:

• Exploring various knowledge in relation extraction (http://delivery.acm.org/10.1145/1220000/1219893/p427-zhou.pdf)

Distant-supervision:

• Learning Syntactic Patterns for Automatic Hypernym Discovery (http://papers.nips.cc/paper/2659-learning-syntactic-patterns-for-automatic-hypernym-discovery.pdf)
• Distant supervision for relation extraction without labeled data (http://www.aclweb.org/anthology/P/P09/P09-1113.pdf)

Unsupervised:

• Open Information Extraction from the Web (http://www.aaai.org/Papers/IJCAI/2007/IJCAI07-429.pdf)
• A Probabilistic Model of Redundancy in Information Extraction (The scoring model used in TextRunner)
• Identifying Relations for Open Information Extraction (http://aclweb.org/anthology/D/D11/D11-1142.pdf)
• The tradeoffs between open and traditional relation extraction.(http://www.kdd.cis.ksu.edu/Courses/Spring-2013/CIS798/Handouts/07-banko08tradeoffs.pdf)
• Unsupervised Relation Discovery with Sense Disambiguation (http://aclweb.org/anthology/P/P12/P12-1075.pdf)