SIG ODIS - Artificial Intelligence and Semantic Technologies for Intelligent Systems
Loading...
Paper Type
Complete
Paper Number
1521
Description
In this information age harmful information and cyberbullying on the Internet have attracted much attention. Most of the existing research on harmful information detection only make breakthroughs in algorithmic efficiency and accuracy, while regarding Internet users as an average whole. However, there could be substantial differences among the users, which to a certain extent, account for the minorities in marginalized groups. Marginalized groups are more likely to be victims of cyberattacks with harmful information, and these cyberattacks often cause more harm to them than others, which creates a vicious cycle. In this paper, we propose a framework for a personalized harmful information detection system based on user portraits. This detection system uses different thresholds to filter harmful information according to a user's personality characteristics, so that we could maximize both the Internet experience and protection of the vulnerable minorities from harmful information.
Recommended Citation
Li, Yuming; Chan, Johnny; Peko, Gabrielle; and Sundaram, David, "A Personalized Harmful Information Detection System Based on User Portraits" (2022). AMCIS 2022 Proceedings. 16.
https://aisel.aisnet.org/amcis2022/sig_odis/sig_odis/16
A Personalized Harmful Information Detection System Based on User Portraits
In this information age harmful information and cyberbullying on the Internet have attracted much attention. Most of the existing research on harmful information detection only make breakthroughs in algorithmic efficiency and accuracy, while regarding Internet users as an average whole. However, there could be substantial differences among the users, which to a certain extent, account for the minorities in marginalized groups. Marginalized groups are more likely to be victims of cyberattacks with harmful information, and these cyberattacks often cause more harm to them than others, which creates a vicious cycle. In this paper, we propose a framework for a personalized harmful information detection system based on user portraits. This detection system uses different thresholds to filter harmful information according to a user's personality characteristics, so that we could maximize both the Internet experience and protection of the vulnerable minorities from harmful information.
When commenting on articles, please be friendly, welcoming, respectful and abide by the AIS eLibrary Discussion Thread Code of Conduct posted here.
Comments
SIG ODIS