# Topic Modeling of Social Media Discourse of Autism Support Groups

**Authors:** Yu Deng, Lei Yang, Juanjuan Chen

PMC · DOI: 10.3390/bs16020280 · Behavioral Sciences · 2026-02-15

## TL;DR

This study uses topic modeling to analyze autism support group discussions on a Chinese social media platform, revealing key themes and sociocultural influences.

## Contribution

The novel application of LDA topic modeling to Chinese autism support group discourse highlights sociocultural impacts and community needs.

## Key findings

- Autism support discussions on Baidu Tieba center on four themes: intervention, education, symptom detection, and community support.
- Sociocultural factors like collectivism and family values strongly influence autism-related discourse and support networks.

## Abstract

Social media platforms serve as critical channels for autism support groups to communicate and seek assistance. This study employed Latent Dirichlet Allocation (LDA) topic modeling to analyze discourse patterns within the Autism Bar on Baidu Tieba, a major Chinese social media. A dataset of 14,151 posts was collected through web crawling, with 12,667 posts retained after preprocessing. The analysis revealed two key findings: (1) The discourse among autism support communities on Baidu Tieba focuses on four central themes: intervention and therapy, early educational journey, early symptom detection and family interaction, and access to educational resources and community support. (2) Sociocultural factors exert a significant influence on autism-related discourse, particularly in shaping societal attitudes toward individuals with autism and the formation of support networks. Traditional Chinese cultural values, such as collectivism and familial centrality, impact the behavioral patterns and decision-making processes of families with autistic children. This study has demonstrated the unique needs and challenges faced by the autism support community, while also informing strategies to promote social media platforms as spaces for support and information exchange. The findings have practical implications for designing targeted interventions and support mechanisms for individuals with autism and their families.

## Linked entities

- **Diseases:** autism (MONDO:0005260)

## Full-text entities

- **Diseases:** Autism (MESH:D001321), language delays (MESH:D007805), anxiety (MESH:D001007), behaviors (MESH:D001523), COVID-19 (MESH:D000086382), ASD (MESH:D000067877), Asperger's syndrome (MESH:D020817), loneliness disease (MESH:D004194), injury to (MESH:D014947), neurodevelopmental condition (MESH:D020763), LDA (MESH:D000085343), disability (MESH:D009069), depression (MESH:D003866)
- **Species:** Homo sapiens (human, species) [taxon 9606]

## Full text

_Full body text omitted from this summary view._ Fetch the complete paper as Markdown: https://tomesphere.com/paper/PMC12937865/full.md

## Figures

5 figures with captions in the complete paper: https://tomesphere.com/paper/PMC12937865/full.md

## References

60 references — full list in the complete paper: https://tomesphere.com/paper/PMC12937865/full.md

---
Source: https://tomesphere.com/paper/PMC12937865