Event Extraction Alone Is Not Enough

Junbo Huang; Longquan Jiang; Cedric Möller; Ricardo Usbeck

Event Extraction Alone Is Not Enough

Junbo Huang^*
, Longquan Jiang
, Cedric Möller
, Ricardo Usbeck^*

^*Korrespondierende/r Autor/-in für diese Arbeit

Professur für Wirtschaftsinformatik, insbesondere Künstliche Intelligenz und Erklärbarkeit

Universität Hamburg

Publikation: Beiträge in Sammelwerken › Konferenzbeitrag › Begutachtung

Abstract

With the growing amount of online data, distinguishing between similar events and news about them poses a significant challenge for both companies and crisis reaction units. To discriminate event instances, we present Eventist, a silver-standard event instance dataset from news in English, containing 23,304 news headlines from 90 countries covering in total 113 storm-related events between 1^st January 2021 and 1^st September 2023. Sampled data is validated by two human raters. Additionally, we propose to adopt a sentence-level event representation for modeling media narrative discourse. Finally, we provide two pairwise comparison benchmarks on event deduplication and event temporal ordering, enabling the practicality of event extraction.

Originalsprache	Englisch
Titel	Narrative Extraction From Texts 2024 : Proceedings of Text2Story — Seventh Workshop on Narrative Extraction From Texts held in conjunction with the 46th European Conference on Information Retrieval (ECIR 2024)
Redakteure/-innen	Ricardo Campos, Alípio Mário Jorge, Adam Jatowt, Simut Bhatia, Marina Litvak
Seitenumfang	10
Band	3671
Erscheinungsort	Aachen
Herausgeber (Verlag)	Rheinisch-Westfaelische Technische Hochschule Aachen
Erscheinungsdatum	05.2024
Seiten	105-114
Publikationsstatus	Erschienen - 05.2024
Veranstaltung	7th Workshop on Narrative Extraction From Texts, Text2Story 2024 - Glasgow, Großbritannien / Vereinigtes Königreich Dauer: 24.03.2024 → 24.03.2024 Konferenznummer: 7 https://text2story24.inesctec.pt/

Bibliographische Notiz

Publisher Copyright:
© 2024 Copyright for this paper by its authors.

Fachgebiete und Schlagwörter

Informatik

Zugriff auf Dokument

Dieses zitieren

Huang, J., Jiang, L., Möller, C., & Usbeck, R. (2024). Event Extraction Alone Is Not Enough. in R. Campos, A. M. Jorge, A. Jatowt, S. Bhatia, & M. Litvak (Hrsg.), Narrative Extraction From Texts 2024: Proceedings of Text2Story — Seventh Workshop on Narrative Extraction From Texts held in conjunction with the 46th European Conference on Information Retrieval (ECIR 2024) (Band 3671, S. 105-114). (CEUR Workshop Proceedings; Band 3671). Rheinisch-Westfaelische Technische Hochschule Aachen. https://ceur-ws.org/Vol-3671/paper9.pdf

Huang, Junbo ; Jiang, Longquan ; Möller, Cedric et al. / Event Extraction Alone Is Not Enough. Narrative Extraction From Texts 2024: Proceedings of Text2Story — Seventh Workshop on Narrative Extraction From Texts held in conjunction with the 46th European Conference on Information Retrieval (ECIR 2024). Hrsg. / Ricardo Campos ; Alípio Mário Jorge ; Adam Jatowt ; Simut Bhatia ; Marina Litvak. Band 3671 Aachen : Rheinisch-Westfaelische Technische Hochschule Aachen, 2024. S. 105-114 (CEUR Workshop Proceedings).

@inproceedings{86a4065c1774405490ac51f228f61240,

title = "Event Extraction Alone Is Not Enough",

abstract = "With the growing amount of online data, distinguishing between similar events and news about them poses a significant challenge for both companies and crisis reaction units. To discriminate event instances, we present Eventist, a silver-standard event instance dataset from news in English, containing 23,304 news headlines from 90 countries covering in total 113 storm-related events between 1st January 2021 and 1st September 2023. Sampled data is validated by two human raters. Additionally, we propose to adopt a sentence-level event representation for modeling media narrative discourse. Finally, we provide two pairwise comparison benchmarks on event deduplication and event temporal ordering, enabling the practicality of event extraction.",

keywords = "Event Deduplication, Event Temporal Ordering, Media Narrative Discourse, Informatics",

author = "Junbo Huang and Longquan Jiang and Cedric M{\"o}ller and Ricardo Usbeck",

note = "Funding Information: The authors acknowledge the financial support by the Federal Ministry for Economic Affairs and Energy of Germany in the project CoyPu (project number 01MK21007G). Publisher Copyright: {\textcopyright} 2024 Copyright for this paper by its authors.; 7th Workshop on Narrative Extraction From Texts, Text2Story 2024 ; Conference date: 24-03-2024 Through 24-03-2024",

year = "2024",

month = may,

language = "English",

volume = "3671",

series = "CEUR Workshop Proceedings",

publisher = "Rheinisch-Westfaelische Technische Hochschule Aachen",

pages = "105--114",

editor = "Ricardo Campos and Jorge, \{Al{\'i}pio M{\'a}rio\} and Adam Jatowt and Bhatia, \{Simut \} and Marina Litvak",

booktitle = "Narrative Extraction From Texts 2024",

address = "Germany",

url = "https://text2story24.inesctec.pt/",

}

Huang, J, Jiang, L, Möller, C & Usbeck, R 2024, Event Extraction Alone Is Not Enough. in R Campos, AM Jorge, A Jatowt, S Bhatia & M Litvak (Hrsg.), Narrative Extraction From Texts 2024: Proceedings of Text2Story — Seventh Workshop on Narrative Extraction From Texts held in conjunction with the 46th European Conference on Information Retrieval (ECIR 2024). Bd. 3671, CEUR Workshop Proceedings, Bd. 3671, Rheinisch-Westfaelische Technische Hochschule Aachen, Aachen, S. 105-114, 7th Workshop on Narrative Extraction From Texts, Text2Story 2024, Glasgow, Großbritannien / Vereinigtes Königreich, 24.03.24. <https://ceur-ws.org/Vol-3671/paper9.pdf>

Event Extraction Alone Is Not Enough. / Huang, Junbo; Jiang, Longquan; Möller, Cedric et al.
Narrative Extraction From Texts 2024: Proceedings of Text2Story — Seventh Workshop on Narrative Extraction From Texts held in conjunction with the 46th European Conference on Information Retrieval (ECIR 2024). Hrsg. / Ricardo Campos; Alípio Mário Jorge; Adam Jatowt; Simut Bhatia; Marina Litvak. Band 3671 Aachen: Rheinisch-Westfaelische Technische Hochschule Aachen, 2024. S. 105-114 (CEUR Workshop Proceedings; Band 3671).

Publikation: Beiträge in Sammelwerken › Konferenzbeitrag › Begutachtung

TY - GEN

T1 - Event Extraction Alone Is Not Enough

AU - Huang, Junbo

AU - Jiang, Longquan

AU - Möller, Cedric

AU - Usbeck, Ricardo

N1 - Conference code: 7

PY - 2024/5

Y1 - 2024/5

N2 - With the growing amount of online data, distinguishing between similar events and news about them poses a significant challenge for both companies and crisis reaction units. To discriminate event instances, we present Eventist, a silver-standard event instance dataset from news in English, containing 23,304 news headlines from 90 countries covering in total 113 storm-related events between 1st January 2021 and 1st September 2023. Sampled data is validated by two human raters. Additionally, we propose to adopt a sentence-level event representation for modeling media narrative discourse. Finally, we provide two pairwise comparison benchmarks on event deduplication and event temporal ordering, enabling the practicality of event extraction.

AB - With the growing amount of online data, distinguishing between similar events and news about them poses a significant challenge for both companies and crisis reaction units. To discriminate event instances, we present Eventist, a silver-standard event instance dataset from news in English, containing 23,304 news headlines from 90 countries covering in total 113 storm-related events between 1st January 2021 and 1st September 2023. Sampled data is validated by two human raters. Additionally, we propose to adopt a sentence-level event representation for modeling media narrative discourse. Finally, we provide two pairwise comparison benchmarks on event deduplication and event temporal ordering, enabling the practicality of event extraction.

KW - Event Deduplication

KW - Event Temporal Ordering

KW - Media Narrative Discourse

KW - Informatics

UR - https://www.scopus.com/pages/publications/85191978124

M3 - Conference contribution

AN - SCOPUS:85191978124

VL - 3671

T3 - CEUR Workshop Proceedings

SP - 105

EP - 114

BT - Narrative Extraction From Texts 2024

A2 - Campos, Ricardo

A2 - Jorge, Alípio Mário

A2 - Jatowt, Adam

A2 - Bhatia, Simut

A2 - Litvak, Marina

PB - Rheinisch-Westfaelische Technische Hochschule Aachen

CY - Aachen

T2 - 7th Workshop on Narrative Extraction From Texts, Text2Story 2024

Y2 - 24 March 2024 through 24 March 2024

ER -

Huang J, Jiang L, Möller C, Usbeck R. Event Extraction Alone Is Not Enough. in Campos R, Jorge AM, Jatowt A, Bhatia S, Litvak M, Hrsg., Narrative Extraction From Texts 2024: Proceedings of Text2Story — Seventh Workshop on Narrative Extraction From Texts held in conjunction with the 46th European Conference on Information Retrieval (ECIR 2024). Band 3671. Aachen: Rheinisch-Westfaelische Technische Hochschule Aachen. 2024. S. 105-114. (CEUR Workshop Proceedings).

Event Extraction Alone Is Not Enough

Abstract

Bibliographische Notiz

Fachgebiete und Schlagwörter

Zugriff auf Dokument

Fingerprint

Dieses zitieren