Overview of the Authorship Verification Task at PAN 2022

Efstathios Stamatatos; Mike Kestemont; Krzysztof Kredens; Piotr Pezik; Annina Heini; Janek Bevendorff; Benno Stein; Martin Potthast

Overview of the Authorship Verification Task at PAN 2022

Efstathios Stamatatos, Mike Kestemont, Krzysztof Kredens, Piotr Pezik, Annina Heini, Janek Bevendorff, Benno Stein, Martin Potthast

Research output: Contribution to journal › Conference article › peer-review

Abstract

The authorship verification task at PAN 2022 follows the experimental setup of similar shared tasks in the recent past. However, it focuses on a different, and very challenging scenario: given two texts belonging to different discourse types, the task is to determine whether they are written by the same author. Based on a new corpus in English, we provide pairs of texts using four discourse types: essays, emails, text messages, and business memos. The differences in communicative purpose, intended audience, and the level of formality render the cross-discourse-type authorship verification task very hard. We received 7 submissions and evaluated them using the TIRA integrated research architecture, along with two baseline approaches. This paper reviews the submissions and presents a detailed discussion of the evaluation results.

Original language	English
Pages (from-to)	2301-2313
Number of pages	13
Journal	CEUR workshop proceedings
Volume	3180
Early online date	9 Aug 2022
Publication status	Published - 5 Sept 2022
Event	Proceedings of the Working Notes of CLEF 2022 - Conference and Labs of the Evaluation Forum - Bologna, Italy Duration: 5 Sept 2022 → 8 Sept 2022 https://ceur-ws.org/Vol-3180/

Bibliographical note

Access to Document

Stamatatos_2022_VoR
© 2021 Copyright for this paper by its authors. Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0).
Final published version, 1.03 MBLicence: CC BY 4.0

https://ceur-ws.org/Vol-3180/paper-184.pdfLicence: CC BY 4.0

Cite this

@article{67f0d2b0e6fc4226961a0e663ae087f0,

title = "Overview of the Authorship Verification Task at PAN 2022",

abstract = "The authorship verification task at PAN 2022 follows the experimental setup of similar shared tasks in the recent past. However, it focuses on a different, and very challenging scenario: given two texts belonging to different discourse types, the task is to determine whether they are written by the same author. Based on a new corpus in English, we provide pairs of texts using four discourse types: essays, emails, text messages, and business memos. The differences in communicative purpose, intended audience, and the level of formality render the cross-discourse-type authorship verification task very hard. We received 7 submissions and evaluated them using the TIRA integrated research architecture, along with two baseline approaches. This paper reviews the submissions and presents a detailed discussion of the evaluation results.",

author = "Efstathios Stamatatos and Mike Kestemont and Krzysztof Kredens and Piotr Pezik and Annina Heini and Janek Bevendorff and Benno Stein and Martin Potthast",

note = "{\textcopyright} 2021 Copyright for this paper by its authors. Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0). ; Proceedings of the Working Notes of CLEF 2022 - Conference and Labs of the Evaluation Forum, CLEF ; Conference date: 05-09-2022 Through 08-09-2022",

year = "2022",

month = sep,

day = "5",

language = "English",

volume = "3180",

pages = "2301--2313",

journal = "CEUR workshop proceedings",

issn = "1613-0073",

publisher = "CEUR-WS.org",

url = "https://ceur-ws.org/Vol-3180/",

}

TY - JOUR

T1 - Overview of the Authorship Verification Task at PAN 2022

AU - Stamatatos, Efstathios

AU - Kestemont, Mike

AU - Kredens, Krzysztof

AU - Pezik, Piotr

AU - Heini, Annina

AU - Bevendorff, Janek

AU - Stein, Benno

AU - Potthast, Martin

PY - 2022/9/5

Y1 - 2022/9/5

N2 - The authorship verification task at PAN 2022 follows the experimental setup of similar shared tasks in the recent past. However, it focuses on a different, and very challenging scenario: given two texts belonging to different discourse types, the task is to determine whether they are written by the same author. Based on a new corpus in English, we provide pairs of texts using four discourse types: essays, emails, text messages, and business memos. The differences in communicative purpose, intended audience, and the level of formality render the cross-discourse-type authorship verification task very hard. We received 7 submissions and evaluated them using the TIRA integrated research architecture, along with two baseline approaches. This paper reviews the submissions and presents a detailed discussion of the evaluation results.

AB - The authorship verification task at PAN 2022 follows the experimental setup of similar shared tasks in the recent past. However, it focuses on a different, and very challenging scenario: given two texts belonging to different discourse types, the task is to determine whether they are written by the same author. Based on a new corpus in English, we provide pairs of texts using four discourse types: essays, emails, text messages, and business memos. The differences in communicative purpose, intended audience, and the level of formality render the cross-discourse-type authorship verification task very hard. We received 7 submissions and evaluated them using the TIRA integrated research architecture, along with two baseline approaches. This paper reviews the submissions and presents a detailed discussion of the evaluation results.

UR - http://www.scopus.com/inward/record.url?scp=85136949420&partnerID=8YFLogxK

UR - https://ceur-ws.org/

M3 - Conference article

AN - SCOPUS:85136949420

SN - 1613-0073

VL - 3180

SP - 2301

EP - 2313

JO - CEUR workshop proceedings

JF - CEUR workshop proceedings

T2 - Proceedings of the Working Notes of CLEF 2022 - Conference and Labs of the Evaluation Forum

Y2 - 5 September 2022 through 8 September 2022

ER -

Overview of the Authorship Verification Task at PAN 2022

Abstract

Bibliographical note

Access to Document

Other files and links

Fingerprint

Cite this