Artificial Intelligence to Address Cyberbullying, Harassment and Abuse: New Directions in the Midst of Complexity

Milosevic, Tijana; Royen, Kathleen Van; Davis, Brian

doi:10.1007/s42380-022-00117-x

Cited by 10 publications

(5 citation statements)

References 17 publications

(9 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Users can report cyberbullying to platforms first (reactive moderation), but AI is also increasingly used to crawl/screen content before it is reported to platforms in an effort of proactive moderation. This process is detailed in some of the large companies' Transparency Reports, which show the amount or percentage of bullying content that was detected and removed proactively (Milosevic, Van Royen, & Davis, 2022).…”

Section: Background On Cyberbullying Interventions and Ai On Social M...mentioning

confidence: 99%

“…Previous research, however, has not solicited children’s views as to the desirability, perceived effectiveness, and privacy impacts of such interventions, which is what we set out to do in our research. Moreover, we also solicit children’s views on proactive content monitoring or screening in direct messages versus on publicly shared content; and the use of facial recognition for cyberbullying detection, all of which are technically feasible on social media platforms even when there is little clarity from the platforms themselves as to whether and how specifically they are implemented (Gorwa et al, 2020; Milosevic, Van Royen, & Davis, 2022; Verma et al, 2022).…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Effectiveness of Artificial Intelligence–Based Cyberbullying Interventions From Youth Perspective

Milosevic

Verma

Carter

et al. 2023

Social Media + Society

Self Cite

View full text Add to dashboard Cite

Cyberbullying on social media continues to pose a significant problem for children and young people, and platforms increasingly rely on artificial intelligence (AI) to address it. Nonetheless, little is known about children’s views as to the effectiveness of AI-based interventions; and how these interventions impact their rights to protection, privacy, and freedom of expression, as afforded to them in the United Nations Convention on the Rights of the Child (UNCRC), which applies in a digital environment. This study is the first to ask children about their perceptions as to how AI-based proactive content moderation of cyberbullying impacts their rights. We have designed a set of AI-based interventions into cyberbullying that build on proactive content take-down, based on social learning and social norm theories; we solicited children’s views via focus groups and in-depth interviews as to their perceived effectiveness and impact of such interventions on children’s rights (children from Ireland, age 12–17, N = 59). We discuss how youth social norms can limit the effectiveness of the interventions and the need to involve youth in platform decisions regarding AI design.

show abstract

Section: Background On Cyberbullying Interventions and Ai On Social M...mentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Effectiveness of Artificial Intelligence–Based Cyberbullying Interventions From Youth Perspective

Milosevic

Verma

Carter

et al. 2023

Social Media + Society

Self Cite

View full text Add to dashboard Cite

show abstract

“…Concerning available resources, most of the proposed studies rely on "easy to access" data (Facebook, YouTube or Instagram) annotated using binary schemes (e.g. content or behaviours categories as abusive or not abusive) (Milosevic, Van Royen, and Davis, 2022). Whilst Van Hee et al (2015) introduce fine-grained annotation guidelines to improve cyberbullying detection, few datasets were released using a more nuanced description of the type of abuse involved, or the severity of the case; or roles played by those involved.…”

Section: Related Workmentioning

confidence: 99%

Harnessing Bullying Traces to Enhance Bullying Participant Role Identification in Multi-Party Chats

Ollagnier

Cabrio

Villata

2023

FLAIRS

View full text Add to dashboard Cite

As online content continues to grow, so does the spread of online hate, especially on social media. Most research efforts conducted on the task of bullying participant role identification are directed towards social networks such as Twitter and Instagram. However, private instant messaging platforms and channels were pinpointed in recent studies as the most prominent grounds for cyberbullying, especially among teens. Since data collection from major social media platforms is strictly limited, very few studies have investigated this task in a multi-party setting. However, the recent release of resources mimicking online aggression situations that may occur among teens on private instant messaging platforms contributes to filling this gap. In this study, we introduce a full pipeline aiming at automating the identification of bullying participant roles (bully and victim) in multi-party chats. Leveraging pre-trained language models and different learning frameworks, we perform hateful content classification of exchanged messages according to a binary scheme (online hate or no online hate). Then, - from these bullying traces - bullying behavioural cues (repetition and intention to harm) are derived and formalised into a role scoring function. As a result, the proposed pipeline identifies the bully and the victim among chat participants. Evaluated against state-of-the-art methods, the proposed pipeline achieves better performances considering all the datasets and roles to predict. In addition, the error analysis confirms that deriving bullying behavioural cues is beneficial to the task of participant role identification.

show abstract

“…However, the efficacy of these provisions is limited. Social media platforms cannot moderate all toxic comments (Mohammad et al, 2016 ; Cho, 2017 ; Badjatiya et al, 2019 ; Sap et al, 2019 ; Milosevic et al, 2022 ). Linguistic filtering approaches can accidentally result in the biased treatment of discriminated minorities (Badjatiya et al, 2019 ; Sap et al, 2019 ).…”

Section: Introductionmentioning

confidence: 99%

“…Linguistic filtering approaches can accidentally result in the biased treatment of discriminated minorities (Badjatiya et al, 2019 ; Sap et al, 2019 ). Removing toxic comments according to the terms of service only removes explicit expressions; therefore, ambiguous and/or cloaked blatant expressions tend to remain on platforms (Mohammad et al, 2016 ; Cho, 2017 ; Milosevic et al, 2022 ). A study revealed that banning the accounts of offenders would not alter their thinking (Johnson et al, 2019 ).…”

Section: Introductionmentioning

confidence: 99%

Online harassment of Japanese celebrities and influencers

Takano,

Taka,

Ogiue

et al. 2024

Front. Psychol.

View full text Add to dashboard Cite

Famous people, such as celebrities and influencers, are harassed online on a daily basis. Online harassment mentally disturbs them and negatively affects society. However, limited studies have been conducted on the online harassment victimization of famous people, and its effects remain unclear. We surveyed Japanese famous people (N = 213), who were influential people who appeared on television and other traditional media and on social media, regarding online harassment victimization, emotional injury, and action against offenders and revealed that various forms of online harassment are prevalent. Some victims used the anti-harassment functions provided by weblogs and social media systems (e.g., blocking/muting/reporting offender accounts and closing comment forms), talked about their victimization to close people, and contacted relevant authorities to take legal action (talent agencies, legal consultants, and police). By contrast, some victims felt compelled to accept harassment and did not initiate action for offenses. We propose several approaches to support victims, inhibit online harassment, and educate people. Our findings help that platforms establish support systems against online harassment.

show abstract

Artificial Intelligence to Address Cyberbullying, Harassment and Abuse: New Directions in the Midst of Complexity

Cited by 10 publications

References 17 publications

Effectiveness of Artificial Intelligence–Based Cyberbullying Interventions From Youth Perspective

Effectiveness of Artificial Intelligence–Based Cyberbullying Interventions From Youth Perspective

Harnessing Bullying Traces to Enhance Bullying Participant Role Identification in Multi-Party Chats

Online harassment of Japanese celebrities and influencers

Contact Info

Product

Resources

About