应用于洪涝灾害案例推理中的层次熵分析法①-计算机系统应用.pdfVIP

计 算 机 系 统 应 用 2012 年 第 2 1 卷 第 4 期 应用于洪涝灾害案例推理中的层次熵分析法① 1,2 1 1,2 文 刚 ，乌 云 ，王儒敬 1( 中国科学院 合肥机械智能研究所，合肥 230031) 2( 中国科学技术大学 自动化系，合肥 230039) 摘 要：案例推理中影响匹配案例检索的关键因素之一就是特征属性权值的分配与确定。将层次熵分析法应用 于洪涝灾害案例推理中，对灾后救灾口粮的需求进行了预估。由层次分析法分析决策问题的目标及多种层次因 素，构造出合理的层次结构。结合主观判断信息得出特征属性权值的优先级。在此基础上利用熵技术对其进行 修正，确保特征属性权值具有客观性，使得优选案例更加切合实际情况。通过实验结果证明，该方法适用于洪 涝灾害案例推理，可以从多项历史案例中选择与目标案例匹配最佳的案例，从而为准确预测救灾口粮需求提供 了科学的参考。 关键词：案例推理；层次分析法；信息熵；特征属性权值 Application of AHP and Entropy to the Case Based Reasoning of Flood Disasters 1,2 1 1,2 WEN Gang , WU Yun , WANG Ru-Jing 1(Department of Automation, University of Science and Technology of China, Hefei 230031, China) 2(Institute of Intelligent Machine, Chinese Academy of Sciences, Hefei 230039, China) Abstract: The key factor that affects the case retrieval matching in Case-based reasoning is the distribution and determination of characteristic attribute weights. In this paper, AHP and Entropy was applied in the Case-based reasoning on flood disaster to estimate the need for disaster relief food rations. AHP was used to analyze and determine the variety of factors which lead to construct a reasonable hierarchy, and get the priority of attribute weights through combining the information of subjective judgments. On this basis, using the entropy technology to modify the priority, and ensure the objectivity of characteristic attribute weights to make the preferred cases fit the reality better. Experimental results show that this method can be applied to the case-based reasoning of floods disaster, and through this method, the best case for the target case can be selected from a number of historical cases. In this way, the scientific estimation of the needs for disaster relief rations can be guaranteed. Key words: case based reasoning; AHP; entropy te