基于论证的智能群决策支持系统

发布时间：2018-03-30 19:22

本文选题：群体决策　切入点：文本情感分析　出处：《湖南科技大学》2015年硕士论文

【摘要】：决策是人们在生活中进行政治、经济活动遇到的在多个可行方案中选择一个最合理的方案的行为。由于个人掌握信息的不全面性,仅依靠个人的智慧进行决策是不合理的。群体决策依靠群体的智慧对问题进行决策,增加了决策结果的可靠性,信服力以及可执行性。群体决策支持系统(GDSS)是在群体决策过程中为决策提供支持的系统。智能决策支持系统结合了人工智能(Artificial Intelligence)与决策支持系统(DSS),使DSS能充分运用人类的智慧,通过逻辑推理来帮助解决复杂的决策问题。由于机器学习,数据挖掘,自然语言处理等人工智能技术的研究不断深入,更多的智能化方法被应用到决策支持系统中,决策支持系统的智能化程度得到明显提高。虽然智能决策支持系统的研究和应用取得较大进展,但仍然存在一些问题,其中包括:1.群体决策过程中决策者的观点一般是基于自然语言的,不便于量化和计算;2.群体决策的过程一般没有可视化展示;3.对方案的支持或者反对力度的计算不够科学;4.较少考虑如何对群体决策过程中观点的分歧和矛盾进行消除。为此,我们需要对群体决策者的决策意见量化、决策过程可视化、计算方案支持或反对力度智能化、以及消除决策者的意见分岐和矛盾进行研究。本文在详细分析和研究目前智能群体决策支持系统的研究现状以及存在的问题基础上,主要做了以下工作:1.针对智能群体决策中决策者的观点不能智能的量化问题,本文提出了利用自然语言处理等技术对决策者观点进行智能分析,推理出决策者的意见。2.针对智能群体决策中决策者的意见量化不够精确的问题,本文运用了模糊推理对群体决策者的意见进行精确的量化,并且考虑了决策者之间的亲密和疏远关系以及决策者的权重值,使决策的结果更加准确同时增加了决策结果的可信度。3.利用百度的Echarts对方案的决策过程等进行可视化展示。利用自然语言处理决策者的观点不仅能使决策者的文本观点进行量化计算,而且提高了决策的智能化程度。文中采用的模糊推理对决策者的意见进行量化计算,提高了方案支持或者反对力度的量化精度。在计算决策者意见时考虑了决策者的权重,更加符合决策的实际。让决策者间亲密、疏远关系参与决策方案的支持或反对力度计算,极大的减少了决策结果受决策者情感因素的影响可能性。可视化决策的整个过程可以使决策者们更加清楚地了解决策结果产生的整个过程。
[Abstract]:Decision making is the behavior that people encounter in political and economic activities in life to choose one of the most reasonable options in a number of feasible schemes. Because of the incompleteness of personal information, It is unreasonable to rely only on the wisdom of the individual to make decisions. Group decision making depends on the wisdom of the group to make decisions on problems, which increases the reliability of the decision results. The Group decision support system (GDSs) is a system that provides support for decision making in the process of group decision making. Intelligent decision support system combines artificial Intelligence with decision support system, which enables DSS to make full use of human intelligence. The research of artificial intelligence technology such as machine learning, data mining, natural language processing and other artificial intelligence technology has been deepened, more intelligent methods have been applied to decision support system, and more and more intelligent methods are applied to decision support system, because of the deep research of artificial intelligence technology, such as machine learning, data mining and natural language processing. Although the research and application of intelligent decision support system has made great progress, there are still some problems. This includes: 1.The views of decision makers in the group decision-making process are generally based on natural language. It is not easy to quantify and calculate. 2. The process of group decision making is generally not visualized. The calculation of support or opposition to the scheme is not scientific enough. 4. Less consideration is given to how to eliminate the differences and contradictions in the process of group decision making. We need to quantify the decision-making opinions of the group decision makers, visualize the decision-making process, and intelligently support or oppose the calculation scheme. On the basis of detailed analysis and research on the current research status and existing problems of intelligent group decision support system, The main work is as follows: 1. Aiming at the problem that the decision maker's viewpoint can not be intelligent in the intelligent group decision making, this paper puts forward the intelligent analysis of the decision maker's viewpoint by using natural language processing and other technologies. To solve the problem that the quantification of the opinion of the decision maker in intelligent group decision-making is not accurate enough, this paper uses fuzzy reasoning to quantify the opinion of the group decision maker accurately. And taking into account the closeness and alienation of decision makers and the weight of decision makers, Making the decision results more accurate and increasing the credibility of the decision results .3.Using Baidu's Echarts to visualize the decision-making process and so on. Using natural language to deal with the views of the decision makers can not only make the text of the decision makers. This view carries out quantitative calculations, The fuzzy reasoning is used to quantify the opinion of the decision maker, and the quantification accuracy of the program support or opposition is improved. The weight of the decision maker is taken into account in the calculation of the decision maker's opinion, and the weight of the decision maker is taken into account in the calculation of the decision maker's opinion. More in line with the realities of decision making. Involve decision makers in the calculation of support or opposition to decision-making programs by closeness, estrangement, It greatly reduces the possibility that the decision results are influenced by the decision makers' emotional factors, and the whole process of visual decision making can make the decision makers understand the whole process of the decision results more clearly.
【学位授予单位】：湖南科技大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：TP311.52

【参考文献】