Detecting Online Risks and Supportive Interaction in Instant
Messenger Conversations using Czech Transformers

D 2021

Detecting Online Risks and Supportive Interaction in Instant Messenger Conversations using Czech Transformers

SOTOLÁŘ, Ondřej; Jaromír PLHÁK; Michal TKACZYK; Michaela LEBEDÍKOVÁ; David ŠMAHEL et al.

Basic information

Original name

Detecting Online Risks and Supportive Interaction in Instant Messenger Conversations using Czech Transformers

Authors

SOTOLÁŘ, Ondřej; Jaromír PLHÁK; Michal TKACZYK ; Michaela LEBEDÍKOVÁ and David ŠMAHEL

Edition

Brno, Recent Advances in Slavonic Natural Language Processing (RASLAN 2021), p. 19-28, 10 pp. 2021

Publisher

Tribun EU

Other information

Language

English

Type of outcome

Proceedings paper

Field of Study

10200 1.2 Computer and information sciences

Country of publisher

Czech Republic

Confidentiality degree

is not subject to a state or trade secret

Publication form

printed version "print"

References:

Domovská stránka workshopu, Full text PDF

Marked to be transferred to RIV

Yes

RIV identification code

RIV/00216224:14330/21:00119420

Organization unit

Faculty of Informatics

ISBN

978-80-263-1670-1

ISSN

EID Scopus

2-s2.0-85126100901

Keywords in English

Online Risks; Supportive Interaction; Facebook Messenger; Text Classification

Changed: 15/5/2024 02:07, RNDr. Pavel Šmerk, Ph.D.

Abstract

In the original language

We present a comparison of state-of-the-art models for text clas- sification of Online Risks and Supportive Interaction in anonymized In- stant Messenger conversations held in Czech. We compare the transformer models Czert, RobeCzech, and FERNET-C5 with the Fasttext classifier as a baseline. For the comparison, we build a novel dataset with five sub- categories for the Online Risks and five for the Supportive Interaction. We solve the balanced classification problem achieving 75.44 - 89.66 F1 score depending on the category. Our results show that the transformer models perform consistently better than the baseline.

Links

GX19-27828X, research and development project

Name: Pohled do budoucnosti: Porozumění vlivu technologií na “well-being” adolescentů (Acronym: FUTURE)

Investor: Czech Science Foundation

Cite

SOTOLÁŘ, Ondřej; Jaromír PLHÁK; Michal TKACZYK; Michaela LEBEDÍKOVÁ and David ŠMAHEL. Detecting Online Risks and Supportive Interaction in Instant Messenger Conversations using Czech Transformers. In Horák, Rychlý, Rambousek. Recent Advances in Slavonic Natural Language Processing (RASLAN 2021). Brno: Tribun EU, 2021, p. 19-28. ISBN 978-80-263-1670-1.

@inproceedings{1809721,
   author = {Sotolář, Ondřej and Plhák, Jaromír and Tkaczyk, Michal and Lebedíková, Michaela and Šmahel, David},
   address = {Brno},
   booktitle = {Recent Advances in Slavonic Natural Language Processing (RASLAN 2021)},
   editor = {Horák, Rychlý, Rambousek},
   keywords = {Online Risks; Supportive Interaction; Facebook Messenger; Text Classification},
   howpublished = {tištěná verze "print"},
   language = {eng},
   location = {Brno},
   isbn = {978-80-263-1670-1},
   pages = {19-28},
   publisher = {Tribun EU},
   title = {Detecting Online Risks and Supportive Interaction in Instant Messenger Conversations using Czech Transformers},
   url = {https://raslan2021.nlp-consulting.net/},
   year = {2021}
}

TY  - CONF
ID  - 1809721
AU  - Sotolář, Ondřej - Plhák, Jaromír - Tkaczyk, Michal - Lebedíková, Michaela - Šmahel, David
PY  - 2021
TI  - Detecting Online Risks and Supportive Interaction in Instant Messenger Conversations using Czech Transformers
PB  - Tribun EU
CY  - Brno
SN  - 9788026316701
KW  - Online Risks
KW  - Supportive Interaction
KW  - Facebook Messenger
KW  - Text Classification
UR  - https://raslan2021.nlp-consulting.net/
N2  - We present a comparison of state-of-the-art models for text clas- sification of Online Risks and Supportive Interaction in anonymized In- stant Messenger conversations held in Czech. We compare the transformer models Czert, RobeCzech, and FERNET-C5 with the Fasttext classifier as a baseline. For the comparison, we build a novel dataset with five sub- categories for the Online Risks and five for the Supportive Interaction. We solve the balanced classification problem achieving 75.44 - 89.66 F1 score depending on the category. Our results show that the transformer models perform consistently better than the baseline.
ER  -

SOTOLÁŘ, Ondřej; Jaromír PLHÁK; Michal TKACZYK; Michaela LEBEDÍKOVÁ and David ŠMAHEL. Detecting Online Risks and Supportive Interaction in Instant Messenger Conversations using Czech Transformers. In Horák, Rychlý, Rambousek. \textit{Recent Advances in Slavonic Natural Language Processing (RASLAN 2021)}. Brno: Tribun EU, 2021, p.~19-28. ISBN~978-80-263-1670-1.

Přehled o publikaci