Privacy Preserving Cyberbullying Prevention with AI Methods in 5G Networks

Abstract

Children and teenagers that have been a victim of bullying can possibly suffer its psychological effects for a lifetime. With the increase of online social media, cyberbullying incidents have been increased as well. In this paper we discuss how we can detect cyberbullying with AI techniques, using term frequencyinverse document frequency. We label messages as benign or bully. We want our method of cyberbullying detection to be privacypreserving, such that the subscribers' benign messages should not be revealed to the operator. Moreover, the operator labels subscribers as normal, bully and victim. The operator utilizes policy control in 5G networks, to protect victims of cyberbullying from harmful traffic.

SoK: Content Moderation for End-to-End Encryption

“…Mis/disinformation [26,118,192,245,248,308] (total: 6) Hate/harassment [237,305,307,384] (total: 4) Nudity/NSFW [281,323] (total: 2) Terrorism & violent extremism [76,212] (total: 2)…”

“…All but three ML-based designs maintained full client privacy; those three were in especially controlled settings of enterprises or parental control [8,304,305]. (Three UK Safety Tech Challenge entries were agnostic as to the client privacy setting [121,162,174].…”

“…A typical setting is to detect or block specific keywords, websites, or content in TLS traffic, usually with no special hardware. MPC or Searchable Encryption [54,55,108,150,182,215,216,226,274,295,304,305,336] (total: 13) Trusted Execution Environment [328] (total: 1)…”

“…This detection approach was overwhelmingly used in TLS middleboxes implementing deep packet inspection via searchable encryption or trusted hardware; see Appendix B for details. Less frequently, this approach was also used for other use cases, including text filtering [269], and detection of spam [34,383], malware and phishing [177,374], and hate speech [305,384].…”

Anonymization Techniques for Privacy Preserving Data Publishing: A Comprehensive Survey

“…Ramezanian et al [219] AI-based privacy preserving system It marks all subscribers as normal at the start, detect the bully message using artificial intelligence (AI) techniques, and protect victims automatically.…”