Scholarlink

<p>The urgent need for climate solutions has spurred innovation at the intersection of data analytics and climate technology. However, a systematic understanding of this emerging startup ecosystem is lacking. This study aims to fill this gap by systematically profiling 5,964 data-driven climate startups. We employ BERTopic, a state-of-the-art natural language processing technique, to analyze company descriptions and patent abstracts from the Net Zero Insights database. Our analysis identifies 38 distinct thematic clusters, yet reveals a stark sectoral imbalance. Commercially mature, mitigation-oriented sectors, such as agriculture and energy monitoring, dominate, while adaptation-critical domains remain marginalized. This thematic skew is mirrored by a geographic concentration in the United States, particularly California, and Europe, which effectively excludes the climate-vulnerable Global South. Financially, the ecosystem displays a “funnel-shaped” dynamic. While abundant seed funding drives early experimentation, the prevalence of acquisition-driven exits suggests a consolidation of technological power within incumbent firms rather than independent scaling. Methodologically, this research demonstrates the utility of advanced NLP for providing a scalable, evidence-based analysis of complex innovation ecosystems, offering a strategic tool to guide investment and policy in climate action.</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

Global sectoral and funding disparities in data-driven climate innovation using BERTopic

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年7月7日

查看全文
  • Pin Pin Tea-makorn,
  • Punyanuch Maitreenukul,
  • Stefano Starita,
  • Pavitra Jindahra

摘要

The urgent need for climate solutions has spurred innovation at the intersection of data analytics and climate technology. However, a systematic understanding of this emerging startup ecosystem is lacking. This study aims to fill this gap by systematically profiling 5,964 data-driven climate startups. We employ BERTopic, a state-of-the-art natural language processing technique, to analyze company descriptions and patent abstracts from the Net Zero Insights database. Our analysis identifies 38 distinct thematic clusters, yet reveals a stark sectoral imbalance. Commercially mature, mitigation-oriented sectors, such as agriculture and energy monitoring, dominate, while adaptation-critical domains remain marginalized. This thematic skew is mirrored by a geographic concentration in the United States, particularly California, and Europe, which effectively excludes the climate-vulnerable Global South. Financially, the ecosystem displays a “funnel-shaped” dynamic. While abundant seed funding drives early experimentation, the prevalence of acquisition-driven exits suggests a consolidation of technological power within incumbent firms rather than independent scaling. Methodologically, this research demonstrates the utility of advanced NLP for providing a scalable, evidence-based analysis of complex innovation ecosystems, offering a strategic tool to guide investment and policy in climate action.