Scholarlink

The Sustainable Development Goals (SDGs) serve as a comprehensive framework for a brighter and more sustainable future. In the development of decision-making systems centered on sustainability, policies play a crucial role in advancing the SDGs and assessing national well-being. Analyzing policy documents and mapping them to relevant Sustainable development goals, targets and indicators is a time-consuming activity that demands significant manual effort from professionals. One of the biggest challenges for this mapping is the existence of a large collection of multi-page, unlabeled policy texts where several SDGs can be mapped to a single document. The existing tools (Bloomberg, KnowSDGs European Commission, GRI SDG Mapping Add-on, UCC SDG Toolkit for Teaching and Learning, and Data4SDGs Toolbox) offer SDG mapping support, however, they do not fully account for policy-making perspectives, including the policy’s intent, implementation, and impact assessment for sustainability. In this paper, we introduce a model that connects policy documents to SDG components (goals, targets, and indicators) through an NLP-driven semantic mapping solution that automates and streamlines the process with the use of topic modeling and sentence embeddings. This analysis supports policymakers by providing insights for crafting policies that align closely with SDGs. The proposed model achieves an accuracy of 90% on a labeled dataset, demonstrating its effectiveness in accurately mapping policy documents to appropriate SDGs.

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

Introspecting Policy Documents Through Semantic Lenses of Sustainable Development Goals

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Apurva Kulkarni,
  • Ifrah Abdul Khadar Ramadurg,
  • Srinath Srinivasa,
  • Sanket Patil

摘要

The Sustainable Development Goals (SDGs) serve as a comprehensive framework for a brighter and more sustainable future. In the development of decision-making systems centered on sustainability, policies play a crucial role in advancing the SDGs and assessing national well-being. Analyzing policy documents and mapping them to relevant Sustainable development goals, targets and indicators is a time-consuming activity that demands significant manual effort from professionals. One of the biggest challenges for this mapping is the existence of a large collection of multi-page, unlabeled policy texts where several SDGs can be mapped to a single document. The existing tools (Bloomberg, KnowSDGs European Commission, GRI SDG Mapping Add-on, UCC SDG Toolkit for Teaching and Learning, and Data4SDGs Toolbox) offer SDG mapping support, however, they do not fully account for policy-making perspectives, including the policy’s intent, implementation, and impact assessment for sustainability. In this paper, we introduce a model that connects policy documents to SDG components (goals, targets, and indicators) through an NLP-driven semantic mapping solution that automates and streamlines the process with the use of topic modeling and sentence embeddings. This analysis supports policymakers by providing insights for crafting policies that align closely with SDGs. The proposed model achieves an accuracy of 90% on a labeled dataset, demonstrating its effectiveness in accurately mapping policy documents to appropriate SDGs.