Story Trees: Representing Documents using Topological Persistence

Pantea Haghighatkhah; Antske Fokkens; Pia Sommerauer; Bettina Speckmann; Kevin Verbeek

Story Trees: Representing Documents using Topological Persistence

Pantea Haghighatkhah, Antske Fokkens, Pia Sommerauer, Bettina Speckmann, Kevin Verbeek

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

Abstract

Topological Data Analysis (TDA) focuses on the inherent shape of (spatial) data. As such, it may provide useful methods to explore spatial representations of linguistic data (embeddings) which have become central in NLP. In this paper we aim to introduce TDA to researchers in language technology. We use TDA to represent document structure as so-called story trees. Story trees are hierarchical representations created from semantic vector representations of sentences via persistent homology. They can be used to identify and clearly visualize prominent components of a story line. We showcase their potential by using story trees to create extractive summaries for news stories.

Original language	English
Title of host publication	2022 Language Resources and Evaluation Conference, LREC 2022
Editors	Nicoletta Calzolari, Frederic Bechet, Philippe Blache, Khalid Choukri, Christopher Cieri, Thierry Declerck, Sara Goggi, Hitoshi Isahara, Bente Maegaard, Joseph Mariani, Helene Mazo, Jan Odijk, Stelios Piperidis
Publisher	European Language Resources Association (ELRA)
Pages	2413-2429
Number of pages	17
ISBN (Electronic)	9791095546726
Publication status	Published - 2022
Event	13th International Conference on Language Resources and Evaluation Conference, LREC 2022 - Marseille, France Duration: 20 Jun 2022 → 25 Jun 2022

Conference

Conference	13th International Conference on Language Resources and Evaluation Conference, LREC 2022
Country/Territory	France
City	Marseille
Period	20/06/22 → 25/06/22

Bibliographical note

Funding Information:
This research is partially funded by the PhD in the Humanities Grant provided by the Netherlands Organization of Scientific Research (Nederlandse Organisatie voor Wetenschappelijk Onderzoek, NWO) PGW.17.041 awarded to Pia Sommerauer. We would like to thank anonymous reviewers for feedback that helped improve this paper. All remaining errors are our own.

Publisher Copyright:
© European Language Resources Association (ELRA), licensed under CC-BY-NC-4.0.

Funding

This research is partially funded by the PhD in the Humanities Grant provided by the Netherlands Organization of Scientific Research (Nederlandse Organisatie voor Wetenschappelijk Onderzoek, NWO) PGW.17.041 awarded to Pia Sommerauer. We would like to thank anonymous reviewers for feedback that helped improve this paper. All remaining errors are our own.

Keywords

Document level discourse
Semantic Vectors
Topical Data Analysis

Cite this

Haghighatkhah, P., Fokkens, A., Sommerauer, P., Speckmann, B., & Verbeek, K. (2022). Story Trees: Representing Documents using Topological Persistence. In N. Calzolari, F. Bechet, P. Blache, K. Choukri, C. Cieri, T. Declerck, S. Goggi, H. Isahara, B. Maegaard, J. Mariani, H. Mazo, J. Odijk, & S. Piperidis (Eds.), 2022 Language Resources and Evaluation Conference, LREC 2022 (pp. 2413-2429). European Language Resources Association (ELRA).

Haghighatkhah, Pantea ; Fokkens, Antske ; Sommerauer, Pia et al. / Story Trees : Representing Documents using Topological Persistence. 2022 Language Resources and Evaluation Conference, LREC 2022. editor / Nicoletta Calzolari ; Frederic Bechet ; Philippe Blache ; Khalid Choukri ; Christopher Cieri ; Thierry Declerck ; Sara Goggi ; Hitoshi Isahara ; Bente Maegaard ; Joseph Mariani ; Helene Mazo ; Jan Odijk ; Stelios Piperidis. European Language Resources Association (ELRA), 2022. pp. 2413-2429

@inproceedings{0eeca591746041ad8113a86882248575,

title = "Story Trees: Representing Documents using Topological Persistence",

abstract = "Topological Data Analysis (TDA) focuses on the inherent shape of (spatial) data. As such, it may provide useful methods to explore spatial representations of linguistic data (embeddings) which have become central in NLP. In this paper we aim to introduce TDA to researchers in language technology. We use TDA to represent document structure as so-called story trees. Story trees are hierarchical representations created from semantic vector representations of sentences via persistent homology. They can be used to identify and clearly visualize prominent components of a story line. We showcase their potential by using story trees to create extractive summaries for news stories.",

keywords = "Document level discourse, Semantic Vectors, Topical Data Analysis",

author = "Pantea Haghighatkhah and Antske Fokkens and Pia Sommerauer and Bettina Speckmann and Kevin Verbeek",

note = "Funding Information: This research is partially funded by the PhD in the Humanities Grant provided by the Netherlands Organization of Scientific Research (Nederlandse Organisatie voor Wetenschappelijk Onderzoek, NWO) PGW.17.041 awarded to Pia Sommerauer. We would like to thank anonymous reviewers for feedback that helped improve this paper. All remaining errors are our own. Publisher Copyright: {\textcopyright} European Language Resources Association (ELRA), licensed under CC-BY-NC-4.0.; 13th International Conference on Language Resources and Evaluation Conference, LREC 2022 ; Conference date: 20-06-2022 Through 25-06-2022",

year = "2022",

language = "English",

pages = "2413--2429",

editor = "Nicoletta Calzolari and Frederic Bechet and Philippe Blache and Khalid Choukri and Christopher Cieri and Thierry Declerck and Sara Goggi and Hitoshi Isahara and Bente Maegaard and Joseph Mariani and Helene Mazo and Jan Odijk and Stelios Piperidis",

booktitle = "2022 Language Resources and Evaluation Conference, LREC 2022",

publisher = "European Language Resources Association (ELRA)",

}

Haghighatkhah, P, Fokkens, A, Sommerauer, P, Speckmann, B & Verbeek, K 2022, Story Trees: Representing Documents using Topological Persistence. in N Calzolari, F Bechet, P Blache, K Choukri, C Cieri, T Declerck, S Goggi, H Isahara, B Maegaard, J Mariani, H Mazo, J Odijk & S Piperidis (eds), 2022 Language Resources and Evaluation Conference, LREC 2022. European Language Resources Association (ELRA), pp. 2413-2429, 13th International Conference on Language Resources and Evaluation Conference, LREC 2022, Marseille, France, 20/06/22.

Story Trees: Representing Documents using Topological Persistence. / Haghighatkhah, Pantea; Fokkens, Antske; Sommerauer, Pia et al.
2022 Language Resources and Evaluation Conference, LREC 2022. ed. / Nicoletta Calzolari; Frederic Bechet; Philippe Blache; Khalid Choukri; Christopher Cieri; Thierry Declerck; Sara Goggi; Hitoshi Isahara; Bente Maegaard; Joseph Mariani; Helene Mazo; Jan Odijk; Stelios Piperidis. European Language Resources Association (ELRA), 2022. p. 2413-2429.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

TY - GEN

T1 - Story Trees

T2 - 13th International Conference on Language Resources and Evaluation Conference, LREC 2022

AU - Haghighatkhah, Pantea

AU - Fokkens, Antske

AU - Sommerauer, Pia

AU - Speckmann, Bettina

AU - Verbeek, Kevin

N1 - Funding Information: This research is partially funded by the PhD in the Humanities Grant provided by the Netherlands Organization of Scientific Research (Nederlandse Organisatie voor Wetenschappelijk Onderzoek, NWO) PGW.17.041 awarded to Pia Sommerauer. We would like to thank anonymous reviewers for feedback that helped improve this paper. All remaining errors are our own. Publisher Copyright: © European Language Resources Association (ELRA), licensed under CC-BY-NC-4.0.

PY - 2022

Y1 - 2022

N2 - Topological Data Analysis (TDA) focuses on the inherent shape of (spatial) data. As such, it may provide useful methods to explore spatial representations of linguistic data (embeddings) which have become central in NLP. In this paper we aim to introduce TDA to researchers in language technology. We use TDA to represent document structure as so-called story trees. Story trees are hierarchical representations created from semantic vector representations of sentences via persistent homology. They can be used to identify and clearly visualize prominent components of a story line. We showcase their potential by using story trees to create extractive summaries for news stories.

AB - Topological Data Analysis (TDA) focuses on the inherent shape of (spatial) data. As such, it may provide useful methods to explore spatial representations of linguistic data (embeddings) which have become central in NLP. In this paper we aim to introduce TDA to researchers in language technology. We use TDA to represent document structure as so-called story trees. Story trees are hierarchical representations created from semantic vector representations of sentences via persistent homology. They can be used to identify and clearly visualize prominent components of a story line. We showcase their potential by using story trees to create extractive summaries for news stories.

KW - Document level discourse

KW - Semantic Vectors

KW - Topical Data Analysis

UR - http://www.scopus.com/inward/record.url?scp=85144248450&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:85144248450

SP - 2413

EP - 2429

BT - 2022 Language Resources and Evaluation Conference, LREC 2022

A2 - Calzolari, Nicoletta

A2 - Bechet, Frederic

A2 - Blache, Philippe

A2 - Choukri, Khalid

A2 - Cieri, Christopher

A2 - Declerck, Thierry

A2 - Goggi, Sara

A2 - Isahara, Hitoshi

A2 - Maegaard, Bente

A2 - Mariani, Joseph

A2 - Mazo, Helene

A2 - Odijk, Jan

A2 - Piperidis, Stelios

PB - European Language Resources Association (ELRA)

Y2 - 20 June 2022 through 25 June 2022

ER -

Haghighatkhah P, Fokkens A, Sommerauer P, Speckmann B , Verbeek K. Story Trees: Representing Documents using Topological Persistence. In Calzolari N, Bechet F, Blache P, Choukri K, Cieri C, Declerck T, Goggi S, Isahara H, Maegaard B, Mariani J, Mazo H, Odijk J, Piperidis S, editors, 2022 Language Resources and Evaluation Conference, LREC 2022. European Language Resources Association (ELRA). 2022. p. 2413-2429

Story Trees: Representing Documents using Topological Persistence

Abstract

Conference

Bibliographical note

Funding

Keywords

Other files and links

Fingerprint

Cite this