Clinical Scorecard: Identification of Cyberbullying Victimization and Assessment Using Large Language Models: Development and Validation of a Cyberbullying Victimization Lexicon

At a Glance

Key Highlights

• Development of a Chinese cyberbullying victimization lexicon with 442 words.
• Strong validity in identifying cyberbullying victimization expressions across three dimensions.
• Large language models showed preliminary utility but require human oversight for complex tasks.
• Cyberbullying victimization is linked to negative mental health outcomes, including anxiety and depression.
• Victims often do not seek help due to shame and fear of retaliation.

Guideline-Based Recommendations

Diagnosis

• Utilize the developed lexicon for identifying expressions of cyberbullying victimization.

Management

• Implement timely psychological interventions for identified victims.

Monitoring & Follow-up

• Assess the effectiveness of interventions based on victim feedback and outcomes.

Risks

• Consider the psychological impact of cyberbullying, including self-harm and suicidal ideation.

Patient & Prescribing Data

Adolescents and individuals affected by cyberbullying.

Early identification and proactive support are critical for mental health outcomes.

Clinical Best Practices

• Adopt a victim-centered definition of cyberbullying for accurate identification.
• Encourage open communication to reduce barriers to seeking help.
• Utilize a collaborative approach between human evaluators and language models in research.

Related Resources & Content

• Zhao et al. (2)
• Hinduja and Patchin (10)
• Alipan et al. (12)