Language-Aware Truth Assessment of Fact Candidates
Nakashole, Ndapandula and Mitchell, Tom M.

Article Structure

Abstract

This paper introduces FactChecker, language-aware approach to truth-finding.

Introduction

Truth-finding algorithms aim to separate true statements (facts) from false information.

Fact Candidates

In this section, we formally define what constitutes a fact candidate and describe how we go about understanding semantics of fact candidates.

Frequent bi- grams

Related Work

There is a fairly small body of work on truth-finding (Yin et al., 2007; Galland et al., 2010; Pasternack and Roth, 2010; Li et al., 2011; Yin and Tan, 2011; Zhao et al., 2012; Pasternack and Roth, 2013).

Conclusion

In this paper, we presented FactChecker, a language-aware approach to truth-finding.

Topics

Mechanical Turk

Appears in 6 sentences as: Mechanical Turk (6)
In Language-Aware Truth Assessment of Fact Candidates
  1. A Mechanical Turk study we carried out revealed that there is a significant correlation between objectivity of language and trustworthiness of sources.
    Page 1, “Introduction”
  2. To test this hypothesis, we designed a Mechanical Turk study.
    Page 2, “Introduction”
  3. (3) Objectivity Classifier: Using labeled data from the Mechanical Turk study, we developed and trained an objectivity classifier which performed better than prior proposed lexicons from literature.
    Page 2, “Introduction”
  4. 3.1 Mechanical Turk Study
    Page 4, “Fact Candidates”
  5. We deployed an annotation study on Amazon Mechanical Turk (MTurk)3, a crowdsourcing platform for tasks requiring human input.
    Page 4, “Fact Candidates”
  6. For training and testing data, we used the labeled data from the Mechanical Turk study.
    Page 4, “Fact Candidates”

See all papers in Proc. ACL 2014 that mention Mechanical Turk.

See all papers in Proc. ACL that mention Mechanical Turk.

Back to top.

entity types

Appears in 5 sentences as: Entity Types (1) entity types (2) entity typing (2)
In Language-Aware Truth Assessment of Fact Candidates
  1. Entity Types .
    Page 2, “Fact Candidates”
  2. We look up entity types in a knowledge
    Page 2, “Fact Candidates”
  3. In particular, we use the NELL entity typing API (Carlson et al., 2010).
    Page 3, “Fact Candidates”
  4. NELL’s entity typing method has high recall because when entities are not in the knowledge base, it performs on-the-fly type inference using the Web.
    Page 3, “Fact Candidates”
  5. Synonymous verbs, relation cardinalities, and entity types enable us to generate alternative fact candidates.
    Page 3, “Fact Candidates”

See all papers in Proc. ACL 2014 that mention entity types.

See all papers in Proc. ACL that mention entity types.

Back to top.

knowledge base

Appears in 5 sentences as: knowledge base (3) knowledge bases (2)
In Language-Aware Truth Assessment of Fact Candidates
  1. These projects have produced knowledge bases containing many millions of relational facts between entities.
    Page 1, “Introduction”
  2. The triple format is the most common representation of facts in knowledge bases .
    Page 2, “Fact Candidates”
  3. NELL’s entity typing method has high recall because when entities are not in the knowledge base , it performs on-the-fly type inference using the Web.
    Page 3, “Fact Candidates”
  4. We evaluated FactChecker on three datasets: i) KB Fact Candidates: The first dataset consists of fact candidates taken from the fact extraction pipeline of a state-of-the-art knowledge base , NELL (Carlson et al., 2010).
    Page 6, “Frequent bi- grams”
  5. ii) Wikipedia Fact Candidates: For the second dataset, we did not restrict the fact candidates to specific topics from a knowledge base , instead we aimed to evaluate all fact candidates about a given entity.
    Page 6, “Frequent bi- grams”

See all papers in Proc. ACL 2014 that mention knowledge base.

See all papers in Proc. ACL that mention knowledge base.

Back to top.

natural language

Appears in 4 sentences as: Natural language (1) natural language (3)
In Language-Aware Truth Assessment of Fact Candidates
  1. Information extraction projects aim to distill relational facts from natural language text (Auer et al., 2007; Bollacker et al,2008;(knlynietal,2010;Faderetal,2011; Nakashole et al., 2011; Del Corro and Gemulla, 2013).
    Page 1, “Introduction”
  2. However, such scores are often tied to the extractor’s ability to read and understand natural language text.
    Page 1, “Introduction”
  3. Natural language is diverse.
    Page 3, “Fact Candidates”
  4. The focus is on understanding natural language , including the use of negation.
    Page 9, “Related Work”

See all papers in Proc. ACL 2014 that mention natural language.

See all papers in Proc. ACL that mention natural language.

Back to top.

News Articles”

Appears in 3 sentences as: news article (1) news articles (2) News Articles” (2)
In Language-Aware Truth Assessment of Fact Candidates
  1. The first task was titled “Trustworthiness of News Articles”, where annotators were given a link to a news article and
    Page 4, “Fact Candidates”
  2. The second task was titled “Objectivity of News Articles” .
    Page 4, “Fact Candidates”
  3. We randomly selected 500 news articles from a corpus of about 300,000 news articles obtained from Google News from the topics of Top News, Business, Entertainment, and SciTech.
    Page 4, “Fact Candidates”

See all papers in Proc. ACL 2014 that mention News Articles”.

See all papers in Proc. ACL that mention News Articles”.

Back to top.

PDF available from www.aclweb.org.
Send us an email