Daffodil International University

Faculty of Science and Information Technology => Recent Technologies and Trends in Software Engineering => Software Engineering => Natural Language Processing => Topic started by: Nazia Nishat on July 13, 2018, 01:01:49 PM

Title: Relation Extraction
Post by: Nazia Nishat on July 13, 2018, 01:01:49 PM

Many NLP applications require understanding relations between word senses: synonymy, antonymy,hyponymy, meronymy. WordNet is machine-readable database of relations between word senses, and an indispensable resource in many NLP tasks.
But WordNet is manually constructed, and has many gaps!
In WordNet 3.1
insulin
progesterone
Not in WordNet 3.1
leptin
pregnenolone

Relation extraction: 5 easy methods
1. Hand-built patterns
2. Bootstrapping methods
3. Supervised methods
4. Distant supervision
5. Unsupervised methods


Source:https://web.stanford.edu/class/cs224u/materials/cs224u-2016-relation-extraction.pdf
Wordnet:http://wordnetweb.princeton.edu/perl/webwn

Title: Re: Relation Extraction
Post by: iftekhar.swe on September 06, 2018, 03:28:08 PM

thanks for sharing

Title: Re: Relation Extraction
Post by: s.arman on April 16, 2019, 07:55:51 PM

Thanks for sharing