病历随访系统中数据挖掘的Apriori算法研究  

Apriori Algorithm Research of Data Mining in Case History and Follow-Up System

作　　者：王卫东[1] 屈洋[2] 

机构地区：[1]暨南大学计算中心,广东广州510632 [2]暨南大学医学院,广东广州510632

出　　处：《计算机技术与发展》2010年第10期4-7,共4页Computer Technology and Development

基　　金：教育部留学回国人员科研启动基金(教外司留[1999]363号)

摘　　要：从存储成千上万份病历的病历随访数据库系统中挖掘出诊断所需的有价值数据,需要掌握有效的挖掘算法实现诊疗方面的数据挖掘。详细论述了数据挖掘的理念和如何根据病历随访数据库内庞大的数据群建立所需的关联规则方法。通过Apriori规则算法分析,建立起目的性极强的数据间的关联规则。通过讨论可以看出选择恰当的关联规则算法不仅可以提高在病历随访数据库中数据挖掘的效率,而且为建立某种疾病的诊疗信息库奠定了基础。Picking up data needed by diagnosis from thousands and thousands of cases stored in case history and follow-up database system,hold a effective mining algorithm to realize data mining of disease treatment.In this study,discuss in detail the theory of data mining and the related practical method of constructing a rule of correlation depending on the mountain of data in the case history and follow-up database.Through Apriori algorithm analysis,constructed a highly purposive correlation rule among the data.Through discussion,see that a suitable Apriori algorithm for the construction of a correlation rule will not only increase the efficiency of data mining in case history and follow-up database,but also set foundation for building a disease treatment information database.

关 键 词：病历随访数据库系统 数据挖掘 Apriori规则算法 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]