Classification of Adolescents' Risky Behavior in Instant
Messaging Conversations

D 2023

Classification of Adolescents' Risky Behavior in Instant Messaging Conversations

PLHÁK, Jaromír, Ondřej SOTOLÁŘ, Michaela LEBEDÍKOVÁ a David ŠMAHEL

Základní údaje

Originální název

Classification of Adolescents' Risky Behavior in Instant Messaging Conversations

Autoři

PLHÁK, Jaromír (203 Česká republika, garant, domácí), Ondřej SOTOLÁŘ (203 Česká republika, domácí), Michaela LEBEDÍKOVÁ (203 Česká republika, domácí) a David ŠMAHEL (203 Česká republika, domácí)

Vydání

https://proceedings.mlr.press, 26th International Conference on Artificial Intelligence and Statistics, AISTATS 2023, od s. 2390-2404, 15 s. 2023

Nakladatel

ML Research Press

Další údaje

Jazyk

angličtina

Typ výsledku

Stať ve sborníku

Obor

10201 Computer sciences, information science, bioinformatics

Utajení

není předmětem státního či obchodního tajemství

Forma vydání

elektronická verze "online"

Odkazy

URL

Kód RIV

RIV/00216224:14330/23:00130551

Organizační jednotka

Fakulta informatiky

ISSN

Klíčová slova anglicky

adolescents; smartphones; machine learning; risky behavior; instant messaging

Štítky

core_A, firank_A

Příznaky

Mezinárodní význam, Recenzováno

Změněno: 8. 4. 2024 16:50, RNDr. Pavel Šmerk, Ph.D.

Anotace

V originále

Previous research on detecting risky online behavior has been rather scattered, typically identifying single risks in online samples. To our knowledge, the presented research is the first that presents a process of building models that can efficiently detect the following four online risky behavior: (1) aggression, harassment, hate; (2) mental health; (3) use of alcohol, and drugs; and (4) sexting. Furthermore, the corpora in this research are unique because of the usage of private instant messaging conversations in the Czech language provided by adolescents. The combination of publicly unavailable and unique data with high-quality annotations of specific psychological phenomena allowed us for precise detection using transformer machine learning models that can handle sequential data and involve the context of utterances. The impact of the context length and text augmentation on model efficiency is discussed in detail. The final model provides promising results with an acceptable F1 score. Therefore, we believe that the model could be used in various applications, e.g., parental applications, chatbots, or services provided by Internet providers. Future research could investigate the usage of the model in other languages.

Návaznosti

GX19-27828X, projekt VaV

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

Investor: Grantová agentura ČR, Modelling the future: Understanding the impact of technology on adolescent’s well-being

MUNI/A/1433/2022, interní kód MU

Název: Zapojení studentů Fakulty informatiky do mezinárodní vědecké komunity 23

Investor: Masarykova univerzita, Zapojení studentů Fakulty informatiky do mezinárodní vědecké komunity 23

Citovat

PLHÁK, Jaromír, Ondřej SOTOLÁŘ, Michaela LEBEDÍKOVÁ a David ŠMAHEL. Classification of Adolescents' Risky Behavior in Instant Messaging Conversations. Online. In Ruiz F., Dy J., van de Meent J-W. 26th International Conference on Artificial Intelligence and Statistics, AISTATS 2023. https://proceedings.mlr.press: ML Research Press, 2023, s. 2390-2404. ISSN 2640-3498.

@inproceedings{2273397,
   author = {Plhák, Jaromír and Sotolář, Ondřej and Lebedíková, Michaela and Šmahel, David},
   address = {https://proceedings.mlr.press},
   booktitle = {26th International Conference on Artificial Intelligence and Statistics, AISTATS 2023},
   editor = {Ruiz F., Dy J., van de Meent J-W.},
   keywords = {adolescents; smartphones; machine learning; risky behavior; instant messaging},
   howpublished = {elektronická verze "online"},
   language = {eng},
   location = {https://proceedings.mlr.press},
   pages = {2390-2404},
   publisher = {ML Research Press},
   title = {Classification of Adolescents' Risky Behavior in Instant Messaging Conversations},
   url = {https://proceedings.mlr.press/v206/plhak23a/plhak23a.pdf},
   year = {2023}
}

TY  - JOUR
ID  - 2273397
AU  - Plhák, Jaromír - Sotolář, Ondřej - Lebedíková, Michaela - Šmahel, David
PY  - 2023
TI  - Classification of Adolescents' Risky Behavior in Instant Messaging Conversations
PB  - ML Research Press
CY  - https://proceedings.mlr.press
KW  - adolescents
KW  - smartphones
KW  - machine learning
KW  - risky behavior
KW  - instant messaging
UR  - https://proceedings.mlr.press/v206/plhak23a/plhak23a.pdf
N2  - Previous research on detecting risky online behavior has been rather scattered, typically identifying single risks in online samples. To our knowledge, the presented research is the first that presents a process of building models that can efficiently detect the following four online risky behavior: (1) aggression, harassment, hate; (2) mental health; (3) use of alcohol, and drugs; and (4) sexting. Furthermore, the corpora in this research are unique because of the usage of private instant messaging conversations in the Czech language provided by adolescents. The combination of publicly unavailable and unique data with high-quality annotations of specific psychological phenomena allowed us for precise detection using transformer machine learning models that can handle sequential data and involve the context of utterances. The impact of the context length and text augmentation on model efficiency is discussed in detail. The final model provides promising results with an acceptable F1 score. Therefore, we believe that the model could be used in various applications, e.g., parental applications, chatbots, or services provided by Internet providers. Future research could investigate the usage of the model in other languages.
ER  -

PLHÁK, Jaromír, Ondřej SOTOLÁŘ, Michaela LEBEDÍKOVÁ a David ŠMAHEL. Classification of Adolescents' Risky Behavior in Instant Messaging Conversations. Online. In Ruiz F., Dy J., van de Meent J-W. \textit{26th International Conference on Artificial Intelligence and Statistics, AISTATS 2023}. https://proceedings.mlr.press: ML Research Press, 2023, s.~2390-2404. ISSN~2640-3498.

Podrobný výpis o publikaci