Scholarlink

<p>While social media is recognized as a key arena for climate discourse in China, the dynamic interplay between narrative framing, public sentiment, and public engagement on platforms like Weibo remains poorly understood. This study employs Latent Dirichlet Allocation (LDA) and a Bidirectional Long Short-Term Memory (BiLSTM) model to track the historical evolution of sentiment across various topics. Furthermore, we examine how narratives vary across media types, interact with public sentiment, and shape communication effectiveness. Results show that the overall discourse remains largely neutral, with limited shifts over time. However, themes of deep reflection and future warnings evoke stronger anger and more dynamic emotional evolution compared to other topics. The discourse on Weibo is characterized by a prevalence of macro-narrative, which are most prominently featured in content from public media. Cross-tabulation further indicates that neutral sentiment is most strongly associated with macro-narrative, whereas pessimistic sentiment is predominantly linked to micro-narratives. Critically, despite their minority presence, micro-narratives generate stronger public engagement than their macro counterparts. This suggests that effective climate communication should strategically embed relatable micro-narrative within authoritative macro-narrative frameworks to simultaneously enhance reach and resonance.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Climate change discourse on Chinese social media: how different narratives shape our emotional connection to the crisis

  • 期刊论文

  • 发表日期：

    2026年1月24日

查看全文
  • Yu Feng,
  • Liuna Geng,
  • Yixin Wang

摘要

While social media is recognized as a key arena for climate discourse in China, the dynamic interplay between narrative framing, public sentiment, and public engagement on platforms like Weibo remains poorly understood. This study employs Latent Dirichlet Allocation (LDA) and a Bidirectional Long Short-Term Memory (BiLSTM) model to track the historical evolution of sentiment across various topics. Furthermore, we examine how narratives vary across media types, interact with public sentiment, and shape communication effectiveness. Results show that the overall discourse remains largely neutral, with limited shifts over time. However, themes of deep reflection and future warnings evoke stronger anger and more dynamic emotional evolution compared to other topics. The discourse on Weibo is characterized by a prevalence of macro-narrative, which are most prominently featured in content from public media. Cross-tabulation further indicates that neutral sentiment is most strongly associated with macro-narrative, whereas pessimistic sentiment is predominantly linked to micro-narratives. Critically, despite their minority presence, micro-narratives generate stronger public engagement than their macro counterparts. This suggests that effective climate communication should strategically embed relatable micro-narrative within authoritative macro-narrative frameworks to simultaneously enhance reach and resonance.