Knowing what to believe (when you already know something)

Jeff Pasternack, Dan Roth

Research output: Contribution to conferencePaper

Abstract

Although much work in NLP has focused on simply determining what a document means, we also must know whether or not to believe it. Fact-finding algorithms attempt to identify the "truth" among competing claims in a corpus, but fail to take advantage of the user's prior knowledge and presume that truth itself is universal and objective rather than subjective. We introduce a framework for incorporating prior knowledge into any fact finding algorithm, expressing both general "common-sense" reasoning and specific facts already known to the user as first-order logic and translating this into a tractable linear program. As our results show, this approach scales well to even large problems, both reducing error and allowing the system to determine truth respective to the user rather than the majority. Additionally, we introduce three new fact-finding algorithms capable of outperforming existing fact-finders in many of our experiments.

Original languageEnglish (US)
Pages877-885
Number of pages9
StatePublished - Dec 1 2010
Event23rd International Conference on Computational Linguistics, Coling 2010 - Beijing, China
Duration: Aug 23 2010Aug 27 2010

Other

Other23rd International Conference on Computational Linguistics, Coling 2010
CountryChina
CityBeijing
Period8/23/108/27/10

Fingerprint

logic
knowledge
experiment
Experiments
Prior Knowledge
First-order Logic
Experiment
Natural Language Processing
Common Sense
Translating

ASJC Scopus subject areas

  • Language and Linguistics
  • Computational Theory and Mathematics
  • Linguistics and Language

Cite this

Pasternack, J., & Roth, D. (2010). Knowing what to believe (when you already know something). 877-885. Paper presented at 23rd International Conference on Computational Linguistics, Coling 2010, Beijing, China.

Knowing what to believe (when you already know something). / Pasternack, Jeff; Roth, Dan.

2010. 877-885 Paper presented at 23rd International Conference on Computational Linguistics, Coling 2010, Beijing, China.

Research output: Contribution to conferencePaper

Pasternack, J & Roth, D 2010, 'Knowing what to believe (when you already know something)', Paper presented at 23rd International Conference on Computational Linguistics, Coling 2010, Beijing, China, 8/23/10 - 8/27/10 pp. 877-885.
Pasternack J, Roth D. Knowing what to believe (when you already know something). 2010. Paper presented at 23rd International Conference on Computational Linguistics, Coling 2010, Beijing, China.
Pasternack, Jeff ; Roth, Dan. / Knowing what to believe (when you already know something). Paper presented at 23rd International Conference on Computational Linguistics, Coling 2010, Beijing, China.9 p.
@conference{bcd7c132212248f9b192999ab1ddf431,
title = "Knowing what to believe (when you already know something)",
abstract = "Although much work in NLP has focused on simply determining what a document means, we also must know whether or not to believe it. Fact-finding algorithms attempt to identify the {"}truth{"} among competing claims in a corpus, but fail to take advantage of the user's prior knowledge and presume that truth itself is universal and objective rather than subjective. We introduce a framework for incorporating prior knowledge into any fact finding algorithm, expressing both general {"}common-sense{"} reasoning and specific facts already known to the user as first-order logic and translating this into a tractable linear program. As our results show, this approach scales well to even large problems, both reducing error and allowing the system to determine truth respective to the user rather than the majority. Additionally, we introduce three new fact-finding algorithms capable of outperforming existing fact-finders in many of our experiments.",
author = "Jeff Pasternack and Dan Roth",
year = "2010",
month = "12",
day = "1",
language = "English (US)",
pages = "877--885",
note = "23rd International Conference on Computational Linguistics, Coling 2010 ; Conference date: 23-08-2010 Through 27-08-2010",

}

TY - CONF

T1 - Knowing what to believe (when you already know something)

AU - Pasternack, Jeff

AU - Roth, Dan

PY - 2010/12/1

Y1 - 2010/12/1

N2 - Although much work in NLP has focused on simply determining what a document means, we also must know whether or not to believe it. Fact-finding algorithms attempt to identify the "truth" among competing claims in a corpus, but fail to take advantage of the user's prior knowledge and presume that truth itself is universal and objective rather than subjective. We introduce a framework for incorporating prior knowledge into any fact finding algorithm, expressing both general "common-sense" reasoning and specific facts already known to the user as first-order logic and translating this into a tractable linear program. As our results show, this approach scales well to even large problems, both reducing error and allowing the system to determine truth respective to the user rather than the majority. Additionally, we introduce three new fact-finding algorithms capable of outperforming existing fact-finders in many of our experiments.

AB - Although much work in NLP has focused on simply determining what a document means, we also must know whether or not to believe it. Fact-finding algorithms attempt to identify the "truth" among competing claims in a corpus, but fail to take advantage of the user's prior knowledge and presume that truth itself is universal and objective rather than subjective. We introduce a framework for incorporating prior knowledge into any fact finding algorithm, expressing both general "common-sense" reasoning and specific facts already known to the user as first-order logic and translating this into a tractable linear program. As our results show, this approach scales well to even large problems, both reducing error and allowing the system to determine truth respective to the user rather than the majority. Additionally, we introduce three new fact-finding algorithms capable of outperforming existing fact-finders in many of our experiments.

UR - http://www.scopus.com/inward/record.url?scp=79955157646&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=79955157646&partnerID=8YFLogxK

M3 - Paper

AN - SCOPUS:79955157646

SP - 877

EP - 885

ER -