In 2015, United Nations adopted 17 Sustainable Development Goals (SDGs) to be achieved by 2030. Implementing SDGs is an important issue for companies as well as for nations. However, to consistently monitor and evaluate progress towards these goals, major efforts in developing objective indices for measuring the levels of SDGs are essential. Currently almost all SDG studies uniformly use available country-level official statistics to synthesize an SDG index, failing to reflect the feelings and opinions of the general public. In this paper we introduce an SDG Social Index developed by a Natural Language Processing (NLP) algorithm. By using social media text data rather than conventional survey data, this index can measure and evaluate the outspoken feelings and opinions of the public toward the SDGs. Here we produce an SDG Social Index for a global company and statistically assess its effectiveness. This framework for producing an SDG Social Index can easily be extended to make other indices where various text data are available. It can also be easily extended to yield country-specific SDG indices across nations. In Conclusion section we summarize the implications of our approach and discuss limitations and future research areas.

2015年，联合国通过了到2030年要实现的17个可持续发展目标。实现可持续发展目标对公司和国家都是重要的问题。但是，为了持续监控和评估实现这些目标的进展，必须做出重大努力来制定用于衡量可持续发展目标水平的客观指标。当前，几乎所有可持续发展目标研究均统一使用国家一级的官方统计数据来综合可持续发展目标指标，但未能反映出公众的感受和意见。在本文中，我们介绍由自然语言处理（NLP）算法开发的SDG社会索引。通过使用社交媒体文本数据而非常规调查数据，该指数可以衡量和评估公众对可持续发展目标的直言不讳的感受和意见。在这里，我们为一家跨国公司编制了SDG社会指数，并对其进行了统计评估。可以很容易地扩展该用于生成SDG社会索引的框架，以在可以使用各种文本数据的情况下创建其他索引。也可以很容易地扩展它以产生国家间特定国家的SDG指数。在结论部分，我们总结了我们的方法的含义，并讨论了局限性和未来的研究领域。