跨境水资源管理已成为21世纪的重大政治和环境挑战之一，其冲突与合作状态对流域可持续发展、区域安全与稳定至关重要，开展跨境河流水冲突与合作动态研究具有重要的理论和现实意义。本文提出了跨境河流水冲突与合作研究框架，按演绎法和归纳法两种研究范式综述了国内外跨境水冲突与合作的研究，对冲突与合作的状态进行定义和量化，对驱动因素和演化规律进行归纳，指出演绎法与归纳法相结合，解释水冲突与合作动态变化的全球规律是重要的前进方向。完善的水冲突与合作数据库是归纳法研究得以开展的数据基础。大数据时代，依靠人工阅读提取信息、建立水冲突与合作动态数据库的方法已经无法应对数据量的爆炸式增长。本文建立了跨境河流水冲突与合作新闻数据挖掘方法。首先，提出了利用新闻媒体文章追踪跨境水冲突与合作动态的数据集构建方法，并提供实用的数据集通用构建工具，使全球跨境河流水冲突与合作数据集在保证一定相关性和覆盖度的基础上可以快速更新，为进一步的数据挖掘分析提供基础。提供了全球286条跨境河流的定制搜索方法，并构建了60条主要跨境河流数据集用于分析。第二，建立了跨境水冲突与合作的判定标准，使新闻文章中涉及的水冲突与合作状态更加科学合理的定义和量化。第三，构建了跨境水冲突与合作分析的专用情感词典，经过与人工判读结果的相关性检验，机器情感分析为动态研究奠定基础。本文对典型跨境流域开展水冲突与合作的动态分析。在构建的全球跨境流域水冲突与合作数据集基础上，通过情感分析、主题分析与利益相关者分析，归纳出全球的普遍规律；同时对四个重点流域（澜湄流域、尼罗河流域、哥伦比亚河流域、恒河-布拉马普特拉河-梅格纳河流域）展开分析，此四流域按照流域国政治互信水平和流域内国家数量进行选取，在全球范围内具有一定代表性。结合全球规律与四重点流域的特异性，归纳出跨境水合作、水冲突分别的导向因素，以及整体呈现出的规律。
Managing transboundary water resources will become one of the major political and environmental challenges in the 21st century. Its conflict and cooperation processes are crucial to the sustainable development of river basins, regional security and stability, and have important theoretical and practical significance. Fully understanding the dynamics of water conflict and cooperation in transboundary rivers is critical to address the challenges. This study proposes a research framework on water conflict and cooperation in transboundary rivers, summarizes the research on transboundary water conflict and cooperation at domestic and abroad according to two research paradigms-deductive method and inductive method. Driving factors and evolutionary laws are summarized. It is pointed out that the combination of deductive and inductive methods to explain the global laws of dynamic changes in water conflict and cooperation is an critical way to go.A complete database of water conflicts and cooperation is the data basis for inductive research to be carried out. In the era of big data, relying on manual reading to extract information and establish a dynamic database of water conflicts and cooperation has been unable to cope with the explosive growth of data volume. This study establishes a data mining method for transboundary water conflict and cooperative news. First, a data set construction method is proposed to use news media articles tracking the dynamics of transboundary water conflicts and cooperation. A powerful dataset construction toolkit is provided, so that the global transboundary water conflict and cooperation dataset can ensure a certain degree of relevance and coverage. It can be quickly updated, providing a basis for further data mining analysis. This study customized the search method for 286 transboundary rivers globally and constructed a dataset of 60 major transboundary rivers for analysis. Second, the criteria for determining transboundary water conflicts and cooperation have been established. The status of water conflicts and cooperation involved in news articles can be more scientifically and reasonably defined and quantified. Third, an exclusive sentiment lexicon for transboundary water conflict and cooperation analysis is constructed. After the correlation test with the results of manual reading, machine sentiment analysis lays the foundation for dynamic research.This study conducts a dynamic water conflict and cooperation analysis in typical transboundary river basins. Based on the constructed dataset of water conflicts and cooperation in global transboundary river basins, sentiment analysis, topic modeling analysis, and stakeholder analysis were conducted to summarize general global trends; four river basins (Mekong, Nile, Columbia, Ganges-Brahmaputra-Megna) are selected according to the level of political mutual trust among riparian countries and the number of riparian countries, which are representative on a global scale. Combined with the global trends and the specificity of the four typical river basins, the guiding factors of transboundary water cooperation and water conflict are summarized and the overall trends globally.
