Scholarlink

<p>The accelerating global waste crisis demands a paradigm shift toward sustainable, resource-efficient solutions, positioning the circular economy framework at the center of this transformation. This study systematically identifies entrepreneurial opportunities within the trash to cash domain by integrating large-scale social media analysis with expert panel validation. Using advanced machine learning and topic modeling techniques, 5453 preprocessed tweets were classified across the three sustainability pillars: environmental, economic, and social. Multi-label classification achieved an overall accuracy of 87.5% and a weighted F1 score of 0.877, demonstrating robust performance in capturing the multidimensional nature of circular economy discourse. The Biterm Topic Model extracted 22 distinct thematic clusters with high coherence scores: environmental topics (K = 3, Cv = 0.481), economic topics (K = 6, Cv = 0.437), and social topics (K = 13, Cv = 0.491). These computational results revealed a diverse opportunity landscape ranging from digital marketplaces for upcycled goods to community-driven biofuel cooperatives to blockchain-enabled traceability services. A multidisciplinary expert panel comprising 12 specialists validated and refined these findings through a three-round modified Delphi process, producing 15 actionable business models aligned with sustainability imperatives. The findings demonstrate that social media serves as both a catalyst for sustainability-driven innovation and a rich source of real-time market intelligence. This integrated methodology advances the field of circular economy research by combining computational rigor with practical relevance, providing policymakers, entrepreneurs, and practitioners with evidence-based strategies to accelerate the transition toward a more sustainable, equitable, and regenerative economic system.</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

Data driven identification of entrepreneurial opportunities in trash to cash initiatives within the circular economy

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年6月3日

查看全文
  • Mehrdad Maghsoudi,
  • Navid Mohammadi,
  • Reza Kia

摘要

The accelerating global waste crisis demands a paradigm shift toward sustainable, resource-efficient solutions, positioning the circular economy framework at the center of this transformation. This study systematically identifies entrepreneurial opportunities within the trash to cash domain by integrating large-scale social media analysis with expert panel validation. Using advanced machine learning and topic modeling techniques, 5453 preprocessed tweets were classified across the three sustainability pillars: environmental, economic, and social. Multi-label classification achieved an overall accuracy of 87.5% and a weighted F1 score of 0.877, demonstrating robust performance in capturing the multidimensional nature of circular economy discourse. The Biterm Topic Model extracted 22 distinct thematic clusters with high coherence scores: environmental topics (K = 3, Cv = 0.481), economic topics (K = 6, Cv = 0.437), and social topics (K = 13, Cv = 0.491). These computational results revealed a diverse opportunity landscape ranging from digital marketplaces for upcycled goods to community-driven biofuel cooperatives to blockchain-enabled traceability services. A multidisciplinary expert panel comprising 12 specialists validated and refined these findings through a three-round modified Delphi process, producing 15 actionable business models aligned with sustainability imperatives. The findings demonstrate that social media serves as both a catalyst for sustainability-driven innovation and a rich source of real-time market intelligence. This integrated methodology advances the field of circular economy research by combining computational rigor with practical relevance, providing policymakers, entrepreneurs, and practitioners with evidence-based strategies to accelerate the transition toward a more sustainable, equitable, and regenerative economic system.