出生缺陷监测数据中的朴素干预规则挖掘  

Mining Nave Intervention Rules in Birth Defect Data

作　　者：张悦[1] 唐常杰[1] 李川[1] 朱军[2] 曾春秋[1] 唐良[1] 刘显宾[1] 

机构地区：[1]四川大学计算机学院,成都610064 [2]中国出生缺陷监测中心四川大学华西医学院,成都610065

出　　处：《计算机科学与探索》2009年第2期188-197,共10页Journal of Frontiers of Computer Science and Technology

基　　金：国家自然科学基金No.60773169;国家"十一五"科技支撑计划No.2006BAI05A01~~

摘　　要：出生缺陷干预规则挖掘是目前医学界和数据挖掘界共同关注的课题。以出生缺陷数据为背景,研究了朴素干预规则建模,并试图发现某些出生缺陷的可能致因。提出了朴素干预规则模型以及朴素干预规则挖掘算法。实验表明,提出的算法能有效挖掘出围产儿缺陷的致因,并为出生缺陷干预工程的政策制定提供致病因素的最佳状态调整方向。Mining naive intervention rules from birth defects data is current hot topic concerned by both the birth defect research area and data mining fields. Taking birth defects data as background, this study aims the modeling of naive intervention rules and discovering the possible causes of specific birth defects from the rough data. The main contributions include： Proposes a new concept called naive intervention rule （NIR）; Proposes and implements the algorithm to mine NIR; Conducts extensive experiments. The empirical result shows that the newly proposed algorithm successfully discovers the causes of prenatal birth defect, and provides evidences to suggest the redirection for intervention decision to reasonable degree.

关 键 词：朴素干预规则 出生缺陷 变化量 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]