Cross-Corpus Textual Entailement for Sublanguage Analysis in Epidemic Intelligence

authored by: Avaré Stewart, Kerstin Denecke, Wolfgang Nejdl
Abstract: Textual entailment has been recognized as a generic task that captures major semantic inference needs across many natural language processing applications. To date, textual entailment has not been considered in a cross-corpus setting, nor for user generated content. The emergence of Medicine 2.0, has made medical blogs an increasingly accepted source of information; but given the characteristics of blogs (which tend to be noisy and informal; or contain a interspersing of subjective and factual sentences) a potentially large amount of irrelevant information may be present. Considering this potential noise, the overarching problem with respect to information extraction from social media for medical intelligence gathering, is achieving the correct level of sentence filtering - as opposed to document or blog post level. In this paper, we propose an approach to textual entailment which uses the text from one source of user generated content (T text) for sentence-level filtering within a new and less amenable one (H text), when the underlying domain, tasks or semantic information is the same, or overlaps.
Organisation(s): L3S Research Centre
Type: Conference contribution
Pages: 2657-2661
No. of pages: 5
Publication date: 01.01.2010
Publication status: Published
Peer reviewed: Yes
ASJC Scopus subject areas: Education, Library and Information Sciences, Linguistics and Language, Language and Linguistics
Electronic version(s): http://www.lrec-conf.org/proceedings/lrec2010/pdf/881_Paper.pdf (Access: Unknown)

BibTeX

@inproceedings{a4cf0168bd404c27b1100985e77138ea,
title = "Cross-Corpus Textual Entailement for Sublanguage Analysis in Epidemic Intelligence",
abstract = "Textual entailment has been recognized as a generic task that captures major semantic inference needs across many natural language processing applications. To date, textual entailment has not been considered in a cross-corpus setting, nor for user generated content. The emergence of Medicine 2.0, has made medical blogs an increasingly accepted source of information; but given the characteristics of blogs (which tend to be noisy and informal; or contain a interspersing of subjective and factual sentences) a potentially large amount of irrelevant information may be present. Considering this potential noise, the overarching problem with respect to information extraction from social media for medical intelligence gathering, is achieving the correct level of sentence filtering - as opposed to document or blog post level. In this paper, we propose an approach to textual entailment which uses the text from one source of user generated content (T text) for sentence-level filtering within a new and less amenable one (H text), when the underlying domain, tasks or semantic information is the same, or overlaps.",
author = "Avar{\'e} Stewart and Kerstin Denecke and Wolfgang Nejdl",
year = "2010",
month = jan,
day = "1",
language = "English",
series = "Proceedings of the 7th International Conference on Language Resources and Evaluation, LREC 2010",
publisher = "European Language Resources Association (ELRA)",
pages = "2657--2661",
editor = "Daniel Tapias and Irene Russo and Olivier Hamon and Stelios Piperidis and Nicoletta Calzolari and Khalid Choukri and Joseph Mariani and Helene Mazo and Bente Maegaard and Jan Odijk and Mike Rosner",
booktitle = "Proceedings of the 7th International Conference on Language Resources and Evaluation, LREC 2010",
note = "7th International Conference on Language Resources and Evaluation, LREC 2010 ; Conference date: 17-05-2010 Through 23-05-2010",
}