基于Apriori算法的医学分类器研究  

Research on the Medical Classification Based on Apriori Algorithm

作　　者：王华[1] 胡学钢[2] 

机构地区：[1]安徽医科大学计算机中心,安徽合肥230032 [2]合肥工业大学计算机与信息学院,安徽合肥230009

出　　处：《计算机仿真》2009年第4期130-134,共5页Computer Simulation

基　　金：安徽省自然科学基金(050420207);安徽医科大学科研基金资助(2008kjzj05)

摘　　要：基于关联分类的数据挖掘在医学临床上得到了广泛应用,但传统算法在进行医学挖掘时却出现选项集大小的制约、项的位置、关联规模过大等问题。针对传统关联分类算法在医学数据挖掘上的这些不足,提出了一种Apriori算法的优化,算法通过加入最大支持度及最小支持度和项出现位置的约束有效地在算法执行前和算法执行过程中删除了冗余模式和规则,满足医学分类要求,提高了医学分类的有效性,并且通过Apriori-gen函数的改进提高了算法运行效率。最后用仿真实验验证了上述优化措施。Data mining based on associated classification is widely applied to clinical medicine, but the problems of the limitation on the size of candidate set, the emerging place of the item, too big in scale of association rules, appear when using traditional algorithm for medical mining. Aiming at the defects existing in the mining association rules of association classification algorithm in medical data, an improved apriori algorithm is proposed by using maxi- mum and minimum support threshold, constraints of item place. The algorithm deletes the redundant models and rules, meets the demand of the medical diagnosis and improves the effectiveness in the medical field. Experimental results show that the algorithm greatly improves the mining efficiency by the improved function Apriori -gen. Computer simulation results show that the optimization measures are practicable.

关 键 词：关联规则 医学分类器 算法 

分 类 号：TP182[自动化与计算机技术—控制理论与控制工程]