LightGCN Based Prediction Model of Anxiety Risk from Adolescent Social Network Behaviors

Read full text Summarise with VIRAT
Authors: Zhiyi Xu
Cited by: 0
References: 4

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

With the growing prevalence of adolescent social media usage, the relationship between online interaction behaviors and psychological health has become a critical research topic.This paper proposes a LightGCN-based prediction model to assess anxiety risk from social network behaviors.The dataset combines structured features (interaction frequency, network centrality, and temporal activity) with unstructured textual data from online mental health questionnaires.By integrating cross-modal feature fusion, the model captures both the topological and semantic dimensions of user behaviors.Experimental results demonstrate that the proposed model achieves superior prediction accuracy compared with traditional machine learning baselines, offering a new approach for early psychological screening and digital mental health intervention.

Fields of Study & Subjects

Mental Health via Writing Impact of Technology on Adolescents Emotion and Mood Recognition

Publication details

Journal / source
Advances in Social Science, Education and Humanities Research/Advances in social science, education and humanities research
Pages
412–418
Publication date
21 March 2026
Language
EN