Sustainable Development Goals (SDGs) bring together the diverse development community and provide a clear set of development targets for 2030. Given a large number of actors and initiatives related to these goals, there is a need to have a way to accurately and reliably assign text to different input: scientific research, research projects, technological output or documents to specific SDGs. In this paper we present Open Source SDG (OSDG) project and tool which does so by integrating existing research and previous classification into a robust and coherent framework. This integration is based on linking the features from the variety of previous approaches, like ontology items, keywords or features from machine-learning models, to the topics in Microsoft Academic Graph.

中文翻译：

---

OSDG - 按照联合国可持续发展目标 (SDG) 对文本数据进行分类的开源方法

可持续发展目标 (SDG) 将多元化的发展社区聚集在一起，并为 2030 年提供了一套明确的发展目标。 鉴于与这些目标相关的众多参与者和举措，需要有一种方法来准确可靠地分配文本不同的投入：科学研究、研究项目、技术产出或特定可持续发展目标的文件。在本文中，我们介绍了开源 SDG (OSDG) 项目和工具，该项目和工具通过将现有研究和以前的分类整合到一个强大且连贯的框架中来实现。这种集成基于将各种先前方法中的功能（例如本体项、关键字或机器学习模型中的功能）与 Microsoft Academic Graph 中的主题相关联。