YouthCare: Building a Personalized Collaborative Video Censorship Tool to Support Parent-Child Joint Media Engagement

Conversational ChatbotsUniversal & Inclusive DesignEarly Childhood EducatorsFamily Caregivers

Research Background and Issues

  • What problems or challenges did the authors identify?

    • Video platforms often lack design considerations for teenagers, making them susceptible to exposure to inappropriate content such as violence, pornography, and crime, which severely impacts their physical and mental health.
    • Current parental approaches to guiding children's media usage rely heavily on joint media engagement mechanisms. However, these methods depend too much on parents' time, knowledge, and experience, making them cumbersome and time-consuming.
    • Designing a tool for video content review that supports joint participation by parents and children, effectively filters risky content, and promotes the selection of age-appropriate videos remains a challenge.
  • Why is this issue important?

    • Teenagers worldwide are extensively using the internet, with a significant portion dedicated to video consumption. The quality of video content directly affects their learning abilities, social development, and mental health, while also influencing family communication quality and parent-child relationships.
    • Addressing this issue not only ensures online safety for teenagers but also fosters collaboration and interaction between parents and children, strengthening family bonds.
  • Research Motivation and Related Work

    • The research is motivated by leveraging the automatic content analysis capabilities of current multimodal large language models (MLLM) to design an intelligent tool that reduces the burden on parents and optimizes joint media engagement between parents and children.
    • Compared to existing work, this study emphasizes multimodal content analysis and personalized collaboration, aiming to go beyond traditional risk control by providing a platform and tools for negotiation between parents and children.

Solution

  • What methods or solutions did the authors propose?

    • The authors designed a personalized collaborative video review tool named "YouthCare," which integrates multimodal large language models (MLLM) to support parents and children in jointly filtering inappropriate content and selecting suitable content.
    • The tool consists of three main modules: a collaborative configuration module, a personalized video review module, and a feedback report module.
  • What are the innovative aspects of this solution?

    • YouthCare introduces a chatbot as an intelligent mediator, enabling parents and teenagers to collaboratively configure video review preferences through dialogue, thereby eliminating potential misunderstandings and conflicts in direct communication.
    • It provides explainable review feedback based on personalized and general guidelines, addressing the challenges of understanding and transparency in video content review.
    • The tool supports real-time feedback and long-term summaries, helping parents not only review immediate content but also understand teenagers' video consumption behaviors while protecting privacy.
  • What are the implementation steps and key technologies used?

    • Collaborative Configuration Module: Teenagers and parents can express preferences either by directly setting keywords or indirectly selecting videos. The chatbot assists in negotiations to generate "joint preferences."
    • Personalized Video Review Module: Multimodal information fusion technology transforms video content into key features, combining personalized (based on joint preferences) and general guidelines (based on authoritative standards) for video analysis.
    • Feedback Report Module: Generates real-time feedback (suitable/unsuitable videos) and long-term summaries (patterns and trend analysis) based on review results.

Research Outcomes

  • What specific outcomes were achieved?

    • YouthCare demonstrated high efficiency in video content review and parent-child collaboration.
    • Experiments showed that its comprehensive functionality, diverse configuration options, and detailed feedback mechanisms received positive evaluations from participants.
  • What advantages does it have compared to existing solutions?

    • Reduced parental burden: By automating reviews and providing explainable feedback, it reduces the time and effort parents need to invest in content review.
    • Collaborative platform: By introducing a chatbot, it mediates collaboration between parents and teenagers, avoiding conflicts in direct communication.
    • Combination of personalization and standardization: Supports personalized preferences while adhering to general review standards.
  • What were the experimental or evaluation results?

    • In evaluations, parents and children gave average scores of 4.5 or higher to different modules of YouthCare, indicating its effectiveness in functionality and user satisfaction.
    • Experimental results also showed that YouthCare significantly enhanced parental engagement and collaborative interaction between parents and children.
  • Limitations and Future Directions

    • Limitations:
      • Younger teenagers may struggle to effectively use text-based interfaces to express preferences, reducing the efficiency of configuration and feedback processes.
      • Lack of long-term impact evaluation, focusing only on short-term effects.
      • Potential cultural or familial context biases in the model.
    • Future Directions:
      • Enhance multimodal interaction capabilities in the user interface, such as supporting voice input and graphical interfaces.
      • Fine-tune the model with customized data to better adapt to family scenarios and reduce detection biases.
      • Conduct long-term usage evaluations to analyze the system's sustained impact on family relationships.

Through comprehensive analysis, this research not only provides a practical solution for improving the efficiency of joint media engagement but also offers numerous feasible guidelines and methods for designing future parent-child collaboration systems.

Quick Actions

Share

Share this page

ios_share

https://hci.top/en/papers/chi/188651/2025

AdRecommended

Learn AI Coding at CodeNow

open_in_newOpen DOI Link
DOI: https://dl.acm.org/doi/10.1145/3706598.3713360
At a Glance

Paper Snapshot

fact_check
dataset
Source
CHI
calendar_month
Year
2025
emoji_events
Award
No award tagged
group
Authors
8 authors
sell
Subtopics
Conversational Chatbots, Universal & Inclusive Design
work
Professions
Early Childhood Educators, Family Caregivers
article
Content Status
Full text indexed
hub
Related Papers
0 related papers