【摘要】 为解决当前应急决策研究大多侧重于数学模型构建且决策数据偏主观假定，导致客观性、实用性及智能性不足的问题，提出基于深度学习和情感倾向分析的单值中智数智能获取方法，并将其应用于突发事件应急决策。首先，利用Python编程技术对突发事件网络舆情数据进行抓取、预处理、统计分析和可视化，得到量化的单值中智数；其次，利用中智数本身具有的不确定性特点，基于精确函数和信息熵客观确定属性权重；最后，利用案例推理(CBR)方法对备选方案进行排序和择优。研究结果表明：所提方法能够对突发事件网络舆情进行实时监测，可以客观智能获取决策数据从而实现对台风灾害的量化评估。研究结果可为相关部门有效应对突发事件网络舆情提供智能辅助决策支持。更多还原

【Abstract】 In order to solve the problem that most of the current research on emergency decision-making focus on the construction of mathematical models, and the decision data is biased towards subjective assumptions, which leads to the lack of objectivity, practicality and intelligence, an intelligent acquisition method of single-valued neutrosophic number based on deep learning and emotional tendency analysis was proposed and applied to the emergency decision-making.Firstly, the Python programming technology was used to carry out the capture, preprocess, statistical analysis and visualization of online public opinion data of emergencies, and the quantified single-valued neutrosophic number was obtained.Secondly, the attribute weights were determined objectively based on the exact function and information entropy by using the uncertainty characteristic of the neutrosophic number itself.Finally, the case-based reasoning(CBR) method was used to rank and prioritize the alternatives.The results show that the proposed method can monitor the online public opinion of emergencies in real time, obtain the decision data objectively and intelligently, thus realize the quantitative evaluation of typhoon disaster.The research results can provide intelligent auxiliary decision-making support for relevant departments to effectively cope with the online public opinion in emergencies.更多还原