SVM在建设项目方案评价中的应用  

Application of SVMs for Construction Scheme Evaluation

作　　者：朱明放[1] 

机构地区：[1]陕西理工学院计算机系,陕西汉中723003

出　　处：《计算机技术与发展》2008年第3期243-245,共3页Computer Technology and Development

基　　金：陕西理工学院科研基金资助项目(SLG0535)

摘　　要：决策分析是一项生产实践中重要的复杂活动,科学的决策需要考虑众多因素,多指标决策分析研究这一方法。多指标决策解决的是有多个属性指标的有限决策方案的排序问题,其关键是获取决策者的偏好信息,多属性效用函数是解决此问题的常用方法。应用支持向量机的多属性决策方法,对建设项目进行评价,阐述了评价方法,并结合实例给出实验结果。实验表明,与其他决策方法比较,支持向量方法作为评价分析更加有效,它能更大区分各个评价对象,扩大支持向量机的应用范围,丰富多属性决策分析的内容。Decision making is an important and complex activity in daily lives,a good decision needs thinking a lots of factors which affect decisions. Multiple attribute decision making is solved sorting o.f limited decision- making projects which have multiple attributes, it is key step to capture and represent the decision maker＇s preference. Multiple attribute utility function is a common method to solve the decision making projects. This paper applies the support vector machines （SVMs） method for multiple attribute decision making to evaluate the construction scheme, at same time to compare this method with other methods which solve same scheme. The study shows based on support vector machine is a good and efficient method for multiple attribute decision,as it can distinguish all candidates, thus enlarge application of SVMs,enrich contents of multiple attribute decision making.

关 键 词：多属性决策 支持向量机 效用函数 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]