Abstract

A major challenge in automated text analysis is that different words are used for related concepts. Analyzing text at the surface level would treat related concepts (i.e. actors, actions, targets, and victims) as different objects, potentially missing common narrative patterns. Shallow parsers reveal semantic roles of words leading to subject-verb-object triplets. We developed a novel algorithm to extract information from triplets by clustering them into generalized concepts by utilizing syntactic criteria based on common contexts and semantic corpus-based statistical criteria based on "contextual synonyms". We show that generalized concepts representation of text (1) overcomes surface level differences (which arise when different keywords are used for related concepts) without drift, (2) leads to a higher-level semantic network representation of related stories, and (3) when used as features, they yield a significant 36% boost in performance for the story detection task.

Original languageEnglish (US)
Title of host publicationProceedings of the 2015 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2015
PublisherAssociation for Computing Machinery, Inc
Pages942-949
Number of pages8
ISBN (Print)9781450338547
DOIs
StatePublished - Aug 25 2015
EventIEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2015 - Paris, France
Duration: Aug 25 2015Aug 28 2015

Other

OtherIEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2015
CountryFrance
CityParis
Period8/25/158/28/15

Fingerprint

Semantics
Syntactics

ASJC Scopus subject areas

  • Computer Science Applications
  • Computer Networks and Communications

Cite this

Ceran, B., Kedia, N., Corman, S., & Davulcu, H. (2015). Story detection using generalized concepts and relations. In Proceedings of the 2015 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2015 (pp. 942-949). Association for Computing Machinery, Inc. https://doi.org/10.1145/2808797.2809312

Story detection using generalized concepts and relations. / Ceran, Betul; Kedia, Nitesh; Corman, Steven; Davulcu, Hasan.

Proceedings of the 2015 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2015. Association for Computing Machinery, Inc, 2015. p. 942-949.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Ceran, B, Kedia, N, Corman, S & Davulcu, H 2015, Story detection using generalized concepts and relations. in Proceedings of the 2015 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2015. Association for Computing Machinery, Inc, pp. 942-949, IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2015, Paris, France, 8/25/15. https://doi.org/10.1145/2808797.2809312
Ceran B, Kedia N, Corman S, Davulcu H. Story detection using generalized concepts and relations. In Proceedings of the 2015 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2015. Association for Computing Machinery, Inc. 2015. p. 942-949 https://doi.org/10.1145/2808797.2809312
Ceran, Betul ; Kedia, Nitesh ; Corman, Steven ; Davulcu, Hasan. / Story detection using generalized concepts and relations. Proceedings of the 2015 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2015. Association for Computing Machinery, Inc, 2015. pp. 942-949
@inproceedings{38b5b5adfa1149a7bcd1b788250f72de,
title = "Story detection using generalized concepts and relations",
abstract = "A major challenge in automated text analysis is that different words are used for related concepts. Analyzing text at the surface level would treat related concepts (i.e. actors, actions, targets, and victims) as different objects, potentially missing common narrative patterns. Shallow parsers reveal semantic roles of words leading to subject-verb-object triplets. We developed a novel algorithm to extract information from triplets by clustering them into generalized concepts by utilizing syntactic criteria based on common contexts and semantic corpus-based statistical criteria based on {"}contextual synonyms{"}. We show that generalized concepts representation of text (1) overcomes surface level differences (which arise when different keywords are used for related concepts) without drift, (2) leads to a higher-level semantic network representation of related stories, and (3) when used as features, they yield a significant 36{\%} boost in performance for the story detection task.",
author = "Betul Ceran and Nitesh Kedia and Steven Corman and Hasan Davulcu",
year = "2015",
month = "8",
day = "25",
doi = "10.1145/2808797.2809312",
language = "English (US)",
isbn = "9781450338547",
pages = "942--949",
booktitle = "Proceedings of the 2015 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2015",
publisher = "Association for Computing Machinery, Inc",

}

TY - GEN

T1 - Story detection using generalized concepts and relations

AU - Ceran, Betul

AU - Kedia, Nitesh

AU - Corman, Steven

AU - Davulcu, Hasan

PY - 2015/8/25

Y1 - 2015/8/25

N2 - A major challenge in automated text analysis is that different words are used for related concepts. Analyzing text at the surface level would treat related concepts (i.e. actors, actions, targets, and victims) as different objects, potentially missing common narrative patterns. Shallow parsers reveal semantic roles of words leading to subject-verb-object triplets. We developed a novel algorithm to extract information from triplets by clustering them into generalized concepts by utilizing syntactic criteria based on common contexts and semantic corpus-based statistical criteria based on "contextual synonyms". We show that generalized concepts representation of text (1) overcomes surface level differences (which arise when different keywords are used for related concepts) without drift, (2) leads to a higher-level semantic network representation of related stories, and (3) when used as features, they yield a significant 36% boost in performance for the story detection task.

AB - A major challenge in automated text analysis is that different words are used for related concepts. Analyzing text at the surface level would treat related concepts (i.e. actors, actions, targets, and victims) as different objects, potentially missing common narrative patterns. Shallow parsers reveal semantic roles of words leading to subject-verb-object triplets. We developed a novel algorithm to extract information from triplets by clustering them into generalized concepts by utilizing syntactic criteria based on common contexts and semantic corpus-based statistical criteria based on "contextual synonyms". We show that generalized concepts representation of text (1) overcomes surface level differences (which arise when different keywords are used for related concepts) without drift, (2) leads to a higher-level semantic network representation of related stories, and (3) when used as features, they yield a significant 36% boost in performance for the story detection task.

UR - http://www.scopus.com/inward/record.url?scp=84962530669&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84962530669&partnerID=8YFLogxK

U2 - 10.1145/2808797.2809312

DO - 10.1145/2808797.2809312

M3 - Conference contribution

SN - 9781450338547

SP - 942

EP - 949

BT - Proceedings of the 2015 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2015

PB - Association for Computing Machinery, Inc

ER -