基于语料库和Gephi的“一带一路”话语分析
Discourse Analysis of “The Belt and Road” Initiative Based on Corpus and Gephi

十多年来，“一带一路”倡议从愿景到行动发展迅速，并且受到了国内外主流媒体的广泛关注。本研究选取美国主流媒体《纽约时报》自2015年至2023年12月间刊登的21篇关于“一带一路”的报道作为研究样本，运用语料库的方法和可视化软件Gephi，对收集到的文本进行索引行分析和共现分析。本研究发现，《纽约时报》对“一带一路”的话语建构积极、中立与消极并存。本文使用Python对收集的文本进行处理，使用nltk库对文本进行分词、词形还原，并识别词组，去掉其中的停用词。对预处理结果进行词频统计，从中筛选出前100个关键词。对关键词进行共现计算，得到词共现矩阵，再使用Gephi绘图。根据得到的关键词图谱进一步分析《纽约时报》围绕着“一带一路”的报道背后所呈现的国际关系与局势，研究发现这些报道背后隐含了错综复杂的国际关系，中国与非洲、美国、欧洲、印度之间的关系都是这些报道关注的热点。
For more than a decade, the “the Belt and Road” initiative has developed rapidly from vision to action, and has received wide media attention at home and abroad. This study selects 21 reports on the “the Belt and Road” initiative published by The New York Times which is the US mainstream media from 2015 to 2023 as research samples. And then this study uses corpus based methods and visualization software Gephi to conduct concordance analysis and co-occurrence analysis of collected texts. The concordance lines of “the Belt and Road” initiative show that there are positive, neutral and negative meanings implied in these reports. This article uses Python to process the collected text, and uses the nltk library to segment the text, restore word forms, identify phrases, and remove stop words. And then this research conducts word frequency analysis on the pre-processing results and selects the top 100 keywords from them. Next, the research performs co-occurrence calculations on the top 100 keywords to obtain a co-occurrence matrix, and then use Gephi to draw the keyword graph. By further analyzing the keyword graph, it can be found that reports related to “the Belt and Road” initiative on the New York Times imply complicated international relations and situations, and the relations between China and Africa, the United States, Europe, and India are the focus of these reports.