摘要
诊断决策过程本质上为信息的处理过程。由于信息结构的复杂性和采集的局限性使得获取的信息存在缺失、模糊、冗余等不完备现象,从而影响诊断的准确性。为此,对条件属性冗余、部分数据值缺失情形下,如何提高被诊断信息的完备性开展讨论,试图通过问题聚类探寻诊断决策所需的隐含规则,提出信息补齐与属性约简的知识挖掘方法:首先,针对Roustida算法在缺失值处理时存在的局限性进行改进,扩充其在工程实践中的适用范围,使缺损信息趋于完整;然后,利用遗传算法和广义诊断规则推理实现条件属性约简和规则凝练;最后,以质量问题诊断为对象进行了案例研究,验证了不完备信息条件下该方法可以实现以相对较简方式表达问题与情境信息之间的关联关系,挖掘问题发生的隐含规律。
The decision-making process is essentially the information processing.The complexity of the information structure and the limitations of data collection usually lead to incomplete phenomena,such as missing,fuzzy and redundant information,which affects the accuracy of diagnosis.For this reason,discussion on how to improve the completeness of the diagnosed information was conducted,and knowledge mining methods of information completion and attribute reduction were proposed.Firstly,the limitations of Roustida algorithm in processing missing values were improved,so as to expand the scope of its application in complex engineering practice and to make the missing information complete.Then,the conditional attribute reduction and rule extraction were carried out by using the genetic algorithm and the generalized diagnostic rule reasoning respectively.Finally,a case study was carried out on the diagnosis of quality issues.It is verified that this method can be used to express the correlation between the problem and contextual information in a relatively simple way under the condition of incomplete information,and explore the underlying laws of problem occurrence.
作者
李金艳
余忠华
LI Jin-yan;YU Zhong-hua(School of Economics and Management,Jiangsu University of Science and Technology,Zhenjiang 212003,China;School of Mechanical Engineering,Zhejiang University,Hangzhou 310027,China)
出处
《科学技术与工程》
北大核心
2023年第35期15117-15123,共7页
Science Technology and Engineering
基金
江苏省高校哲学社会科学研究基金(2018SJA1090)。
关键词
信息不完备
信息补齐
条件属性冗余
属性约简
规则提取
information incompleteness
information filling
conditional attribute redundancy
attribute reduction
rule extraction
