摘要
作为人际关系和政治体验的直接表达,情感是政治决策中难以排除、也难以回避的扰动因素,但是由于传统国际关系研究理性范式的盛行和情感分析技术的缺位,情感分析作为一种研究范式一直未能受到应有的重视。当前,随着大数据时代的到来,新素材、新技术和新思维的出现使得大数据情感分析有可能发展成为国际关系研究走向工程应用的一个重要实验领域。整体上,大数据情感分析主要立足于计算机自然语言处理技术与国际关系专业知识的跨学科融合,重在探索作为非理性因素的情感对于现代政治运行的隐秘影响。放眼未来,大数据情感分析有着广阔的拓展空间,但其发展也面临若干困境。文章以政治情感为切入点,从政治情感分析的理论演化及技术进展两个方面入手,分析了情感计算的基本原理,探索了大数据时代情感分析技术在国际关系领域内的应用场景,并简要评估了大数据情感分析的优势和不足,旨在捕捉复杂社会中的微妙人际互动和情绪传播。
Does the online buzz among voters before the election indicate their voting preferences?Does the spread of negative emotions such as panic,anger,and hatred herald social unrest?Does the war of words between two countries portend the future direction of bilateral relations?As a direct expression of interpersonal relations and political experience,sentiment is a factor which is difficult to exclude and avoid in political decision-making.However,in international relations,sentiment analysis,as a research paradigm,has not been given due attention due to the prevalence of the traditional rational paradigm and the lack of necessary tools for emotion analysis.Nowadays,with the advent of big data and the emergence of new materials,technologies and ideas,big data-based sentiment analysis is likely to be an important IR subfield to experiment with engineering application.Overall,sentiment analysis is mainly based on the interdisciplinary integration of natural language processing technology in computer science and expertise in international relations.It focuses on exploring the hidden influence of emotions as a non-rational factor on modern politics.While big data-based sentiment analysis has a great potential for expansion,it will also face many difficulties.By looking at the theoretical and technological progress in the analysis of political sentiment,this article analyzes the basic principles and application scenarios of sentiment analysis in IR.It also briefly evaluates the strengths and weaknesses of sentiment analysis in its attempt to capture the subtle human interaction and emotional transmission in a complex society.
作者
王云璇
董青岭
Wang Yunxuan;Dong Qingling
出处
《国际论坛》
CSSCI
北大核心
2020年第6期64-85,157,共23页
International Forum
基金
对外经济贸易大学国家对外开放研究院“构建新型国际关系的中美认知差异与我国推进策略研究”项目、北京市社科基金重大项目“中国在国际舞台上的地位与作用研究(19ZDA14)”之阶段性成果
惠园杰出青年学者项目“基于新闻事件自动编码技术的大数据冲突预测模型研究(18JQ03)”的赞助支持。
关键词
数字时代
情感分析
大数据
国际关系
非理性范式
digital age
sentiment analysis
big data
international relations
nonrational paradigm
