摘要
为较好地解决高层建筑结构方案设计阶段知识发现问题,突破知识获取“瓶颈”.针对信息系统为连续属性的情况,提出了基于粗糙集理论的高层建筑结构方案设计的知识发现方法.通过模糊隶属函数将连续属性值表示成模糊值,利用区分矩阵删除冗余条件属性,通过m ax{|posCi(D)|}值的计算,考查各条件属性对决策分类的重要性,选择涵盖最多决策分类对象的属性为分枝结点,便可得到一棵与最小决策规则集对应的决策树,从而导出决策规则.最后给出一个应用实例.
To solve the problem of knowledge discovery in scheme design of high - rise and to break through the "bottle neck" of knowledge building structures, acquisition, a new method for knowledge discovery based on rough set theory is proposed for continuous attribute decision system. The method transforms the continuous attribute values into the fuzzy ones with fuzzy subjection function by using discernibility matrix to delete redundant condition attributes. The values of max{|posCi (D) |} are calculated to check the weights of each condition attribute for decision - making classification, and then, the attribute that cover most decision - making classification objects as ramification nodes is selected. Then a decision tree that corresponds to the collection of least decision - making rules is found so that the decision rule can be deduced. At last, an application example is presented to illustrate the feasibility and efficiency of the proposed method.
出处
《哈尔滨工业大学学报》
EI
CAS
CSCD
北大核心
2007年第2期177-180,共4页
Journal of Harbin Institute of Technology
基金
国家自然科学基金资助项目(50378030)
黑龙江省科技攻关资助项目(GC05A112)
黑龙江省博士后科研究启动金资助项目(LHK-04072)
关键词
高层建筑
结构方案设计
知识发现
粗糙集
决策树
high-rise building
structural scheme design
knowledge discovery
rough sets
decision tree