Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Domain-Specific Knowledge Graph ConstructionInformation Extraction

Domain-Specific Knowledge Graph Construction: Information Extraction [Extracting information from both Web and natural language documents is the central step in knowledge graph construction, since it is the first line of attack in going from a corpus that is not machine-understandable or queryable to a semi-structured corpus that can be queried and reasoned over. Wrapper induction techniques were developed early in the Web community to deal with the special problem of extracting information from webpages and web templates. However, wrapper induction is not enough. Many key attributes need to be extracted directly from text using information extraction algorithms developed in the natural language processing community. This is also true in cases where the raw data is not from the Web, but is a corpus of natural language documents to begin with. Therefore, we also cover some established research on information extraction, including named entity recognition, relation extraction and event extraction. While the first of these has been around for quite some time, the last is a relatively novel research area where improving quality continues to be a challenge.] http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png

Domain-Specific Knowledge Graph ConstructionInformation Extraction

Download PDF
22 pages

Loading next page...
 
/lp/springer-journals/domain-specific-knowledge-graph-construction-information-extraction-22T50geg1p
Publisher
Springer International Publishing
Copyright
© The Author(s), under exclusive license to Springer Nature Switzerland AG 2019
ISBN
978-3-030-12374-1
Pages
9 –31
DOI
10.1007/978-3-030-12375-8_2
Publisher site
See Chapter on Publisher Site

Abstract

[Extracting information from both Web and natural language documents is the central step in knowledge graph construction, since it is the first line of attack in going from a corpus that is not machine-understandable or queryable to a semi-structured corpus that can be queried and reasoned over. Wrapper induction techniques were developed early in the Web community to deal with the special problem of extracting information from webpages and web templates. However, wrapper induction is not enough. Many key attributes need to be extracted directly from text using information extraction algorithms developed in the natural language processing community. This is also true in cases where the raw data is not from the Web, but is a corpus of natural language documents to begin with. Therefore, we also cover some established research on information extraction, including named entity recognition, relation extraction and event extraction. While the first of these has been around for quite some time, the last is a relatively novel research area where improving quality continues to be a challenge.]

Published: Mar 5, 2019

There are no references for this article.