一个新的最优简单规则挖掘算法.pdfVIP

小细微型计算机赢统 2009 年 3 月第3 期 Journal of Chinese Computer Systems Vol.30 No.3 2009 一个新的最优简单规则挖掘算法 邱卫根1 ， 2 ，徐柏林1 ， 3 I( 广东工业大学计算机学院，广东广州 510090) 气智能技术与系统Lïl家援点实验交.计算机科学与技术系，北京 100084) 气广州工程技术职业学院.广东广州 510090) E-mail;junhuiqiu@163.com.xuxI2012@126.com. 摘 瞿z 关联规则的4是嚣址是知识发现和~秘挖掘锁城中的金奖问趟，粗集双论是研究规则挖掘伪重婆的数学工具.本义研究仿 息准统命运优简单关联统则挖掘.Jr.法.本文利用信急关联短路中元.1世特钱，对Jt.泼仔交换.i4是从命袋就关联线.)潜在份杂伶 元，以此作为缆.IIH也极算法的基础.本文的..J条简单1展现，能挖掘出仿息系统中所有最优简单规则.而且有放池边免，丁通常属性 约简垃圾中伪 NP-hard 问题.本文以{个实例证明本方泌的有效性. 关键满2 最简规则z 程集理诠z 英联短悴 中团分提号: TP18 :fc献掠识码:A 文章辅导:1000-1220(2∞的03-0417翩创 New Algorithm for Optimal Minimal Rules Mining t 1 QIUWeí啄en . 章， XU Xiang-lin •事 I(C酬阳， ter Faculty 01 (ÃωngDong Un;fJersity 01 Ttchnology. 仙刷gll:，加u 510090. China) 1(State Key Lab. ollmelligmt Technology and Sysum. Departm棚tolCOI呻UtI!r Scimce and Tech饰。logy. Tsinghua UnitJerI;协， Btijjn8100084. China) 气也ang%加， u lnllitute 0/ Technolo.即• Guangll:hou 510090 , Clúna) Abstract I Acquisition for associatíon rules of information systems is an ímportant problem oE knowledge treatm曹nt and data míni吨. and ro吨h 掬t8 theory Îs an import器nt mathematical tools of rules míning. T如paper rese革rches the a1gorithm for opti峭 mal minimal mining rules in ínformation systems. By u串e of the characterístíc of elements ín assoeiation matrix. the algorithm proposed in this paper can find the conditional elements of potential assoeiation rules including in the information systems after transforming the property elements of association matri嚣· 丁he algorithm can directly simplify the mining progre串串 of optimal minimal rules of information systems , and effectively void the NP啪hard prob1em often oecurring În the attributes reductÎon of information systems. An example is a1so given to prove