The Belt and Road Initiative on Twitter: An Annotated Dataset

Initiated by the Chinese president Xi Jinping in 2013, the Belt and Road initiative (BRI) is a multi-trillion-dollar agenda for facilitating trade and investment, especially massive infrastructural developments. In recent years, discussions around the BRI have been increasing as more than 130 countries and 30 international organizations have officially joined the initiative, collaborating in a series of transnational infrastructure projects funded by Chinese companies or the Chinese state. This dataset provides 500,711 posts and 714,794 reposting threads related to the BRI on Twitter. The dataset was collected through the Twitter API by applying a set of keywords: “belt and road”, “one belt one road”, “new silk road”, “maritime silk road”, and “silk road economic belt”, which included the words and their hashtag forms to download the raw data from Twitter. The time series of the dataset is from 7 September 2013 to 30 November 2021. Furthermore, the dataset is annotated in terms of languages, emotional polarity, geopolitical entities, and credibility by employing textual analytics in language detection, neural machine translation, and lexicon-based sentiment analysis. To facilitate future research, we classified the dataset into three databases that can be analyzed separately and reused in research related to various fields, such as political science, network science, and sociology to study public opinions about the BRI and their dissemination patterns.

All programming scripts required to reproduce the dataset are available at: https://github.com/edmangog/The-BRI-on-Twitter

由中华人民共和国主席习近平于2013年发起的“一带一路”倡议（Belt and Road Initiative，简称BRI），是一项旨在促进贸易与投资、尤其是大规模基础设施建设的数万亿美元宏伟蓝图。近年来，随着超过130个国家及30个国际组织正式加入该倡议，并在中国企业或中国国家支持下共同参与一系列跨国基础设施项目的合作，围绕“一带一路”的讨论日益增多。本数据集收录了与“一带一路”相关的500,711篇推文及其714,794个转发线程，通过Twitter API，运用关键词“belt and road”、“one belt one road”、“new silk road”、“maritime silk road”和“silk road economic belt”，包括其单词形式和标签形式，下载了Twitter上的原始数据。数据的时间序列跨度为2013年9月7日至2021年11月30日。此外，通过语言检测、神经机器翻译和基于词汇的情感分析等文本分析方法，对数据集进行了语言、情感极性、地缘政治实体和可信度的标注。为便于未来研究，我们将数据集划分为三个可独立分析和在政治学、网络科学和社会学等领域的相关研究中重复使用的数据库，以研究公众对“一带一路”的看法及其传播模式。