Document Cohesion Flow: Striving towards Coherence

Scott Crossly, Mihai Dascalu, Stefan Trausan-Matu, Laura Allen, Danielle S. McNamara

Research output: Chapter in Book/Report/Conference proceedingConference article in proceedingAcademicpeer-review

11 Downloads (Pure)

Abstract

Text cohesion is an important element of discourse
processing. This paper presents a new approach to modeling,
quantifying, and visualizing text cohesion using automated
cohesion flow indices that capture semantic links among
paragraphs. Cohesion flow is calculated by applying
Cohesion Network Analysis, a combination of semantic
distances, Latent Semantic Analysis, and Latent Dirichlet
Allocation, as well as Social Network Analysis. Experiments
performed on 315 timed essays indicated that cohesion flow
indices are significantly correlated with human ratings of text
coherence and essay quality. Visualizations of the global
cohesion indices are also included to support a more facile
understanding of how cohesion flow impacts coherence in
terms of semantic dependencies between paragraphs.
Original languageEnglish
Title of host publicationCogSci 2016 Proceedings
EditorsA. Papafragou, D. Grodner, D. Mirman, J.C. Trueswell
Place of PublicationAustin, TX
PublisherCognitive Science Society
Pages764-769
ISBN (Print)978-0-9911967-3-9
Publication statusPublished - Aug 2016
Externally publishedYes
Event38th Annual Conference of the Cognitive Science Society: Recognizing and Representing Events - Philadelphia, United States
Duration: 10 Aug 201613 Aug 2016
https://mindmodeling.org/cogsci2016/

Conference

Conference38th Annual Conference of the Cognitive Science Society
Abbreviated titleCOGSCI2016
CountryUnited States
CityPhiladelphia
Period10/08/1613/08/16
Internet address

Fingerprint

Semantics
Electric network analysis
Visualization

Keywords

  • Cohesion Flow
  • Natural Language Processing
  • Computational Models
  • Cohesion Network Analysis
  • Coherence
  • Writing Quality

Cite this

Crossly, S., Dascalu, M., Trausan-Matu, S., Allen, L., & McNamara, D. S. (2016). Document Cohesion Flow: Striving towards Coherence. In A. Papafragou, D. Grodner, D. Mirman, & J. C. Trueswell (Eds.), CogSci 2016 Proceedings (pp. 764-769). Austin, TX: Cognitive Science Society.
Crossly, Scott ; Dascalu, Mihai ; Trausan-Matu, Stefan ; Allen, Laura ; McNamara, Danielle S. / Document Cohesion Flow: Striving towards Coherence. CogSci 2016 Proceedings. editor / A. Papafragou ; D. Grodner ; D. Mirman ; J.C. Trueswell. Austin, TX : Cognitive Science Society, 2016. pp. 764-769
@inproceedings{39d2820f41f449ba9c58a3ff5850880d,
title = "Document Cohesion Flow: Striving towards Coherence",
abstract = "Text cohesion is an important element of discourseprocessing. This paper presents a new approach to modeling,quantifying, and visualizing text cohesion using automatedcohesion flow indices that capture semantic links amongparagraphs. Cohesion flow is calculated by applyingCohesion Network Analysis, a combination of semanticdistances, Latent Semantic Analysis, and Latent DirichletAllocation, as well as Social Network Analysis. Experimentsperformed on 315 timed essays indicated that cohesion flowindices are significantly correlated with human ratings of textcoherence and essay quality. Visualizations of the globalcohesion indices are also included to support a more facileunderstanding of how cohesion flow impacts coherence interms of semantic dependencies between paragraphs.",
keywords = "Cohesion Flow, Natural Language Processing, Computational Models, Cohesion Network Analysis, Coherence, Writing Quality",
author = "Scott Crossly and Mihai Dascalu and Stefan Trausan-Matu and Laura Allen and McNamara, {Danielle S.}",
year = "2016",
month = "8",
language = "English",
isbn = "978-0-9911967-3-9",
pages = "764--769",
editor = "A. Papafragou and D. Grodner and D. Mirman and J.C. Trueswell",
booktitle = "CogSci 2016 Proceedings",
publisher = "Cognitive Science Society",
address = "United States",

}

Crossly, S, Dascalu, M, Trausan-Matu, S, Allen, L & McNamara, DS 2016, Document Cohesion Flow: Striving towards Coherence. in A Papafragou, D Grodner, D Mirman & JC Trueswell (eds), CogSci 2016 Proceedings. Cognitive Science Society, Austin, TX, pp. 764-769, 38th Annual Conference of the Cognitive Science Society, Philadelphia, United States, 10/08/16.

Document Cohesion Flow: Striving towards Coherence. / Crossly, Scott; Dascalu, Mihai; Trausan-Matu, Stefan; Allen, Laura; McNamara, Danielle S.

CogSci 2016 Proceedings. ed. / A. Papafragou; D. Grodner; D. Mirman; J.C. Trueswell. Austin, TX : Cognitive Science Society, 2016. p. 764-769.

Research output: Chapter in Book/Report/Conference proceedingConference article in proceedingAcademicpeer-review

TY - GEN

T1 - Document Cohesion Flow: Striving towards Coherence

AU - Crossly, Scott

AU - Dascalu, Mihai

AU - Trausan-Matu, Stefan

AU - Allen, Laura

AU - McNamara, Danielle S.

PY - 2016/8

Y1 - 2016/8

N2 - Text cohesion is an important element of discourseprocessing. This paper presents a new approach to modeling,quantifying, and visualizing text cohesion using automatedcohesion flow indices that capture semantic links amongparagraphs. Cohesion flow is calculated by applyingCohesion Network Analysis, a combination of semanticdistances, Latent Semantic Analysis, and Latent DirichletAllocation, as well as Social Network Analysis. Experimentsperformed on 315 timed essays indicated that cohesion flowindices are significantly correlated with human ratings of textcoherence and essay quality. Visualizations of the globalcohesion indices are also included to support a more facileunderstanding of how cohesion flow impacts coherence interms of semantic dependencies between paragraphs.

AB - Text cohesion is an important element of discourseprocessing. This paper presents a new approach to modeling,quantifying, and visualizing text cohesion using automatedcohesion flow indices that capture semantic links amongparagraphs. Cohesion flow is calculated by applyingCohesion Network Analysis, a combination of semanticdistances, Latent Semantic Analysis, and Latent DirichletAllocation, as well as Social Network Analysis. Experimentsperformed on 315 timed essays indicated that cohesion flowindices are significantly correlated with human ratings of textcoherence and essay quality. Visualizations of the globalcohesion indices are also included to support a more facileunderstanding of how cohesion flow impacts coherence interms of semantic dependencies between paragraphs.

KW - Cohesion Flow

KW - Natural Language Processing

KW - Computational Models

KW - Cohesion Network Analysis

KW - Coherence

KW - Writing Quality

M3 - Conference article in proceeding

SN - 978-0-9911967-3-9

SP - 764

EP - 769

BT - CogSci 2016 Proceedings

A2 - Papafragou, A.

A2 - Grodner, D.

A2 - Mirman, D.

A2 - Trueswell, J.C.

PB - Cognitive Science Society

CY - Austin, TX

ER -

Crossly S, Dascalu M, Trausan-Matu S, Allen L, McNamara DS. Document Cohesion Flow: Striving towards Coherence. In Papafragou A, Grodner D, Mirman D, Trueswell JC, editors, CogSci 2016 Proceedings. Austin, TX: Cognitive Science Society. 2016. p. 764-769