基因的无树评估(英文)  

Evaluating genes without expression trees construction

作　　者：李川[1] 唐常杰[1] 陈瑜[1] 代术成[1] 邱江涛[1] 罗谦[1] 朱军[2] 

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

出　　处：《计算机工程与应用》2008年第14期80-84,共5页Computer Engineering and Applications

基　　金：the National Natural Science Foundation of China under Grant No.90409007(国家自然科学基金);the11th Five Years Key Programs for Sci.&Tech.De-velopment of China under grant No.2006038002003(国家"十一五"科技支撑项目);the Tackle Key Problem Fund of Sichuan Province under Grant No.2006Z01-027(四川省科技攻关项目);Sichuan University Youth Foundation under Grant No.06036(四川大学青年基金)

摘　　要：传统基于表达式树ET的基因评估从性能角度讲主要缺点是:重复遍历表达式树和进行大量重复计算。提出基于Scale的基因评估。变量矩阵用来避免基因评估中的重复计算。实验表明,在绝大多数数据分布下和参数选择情况下:基于Scale的基因评估较基于ET的基因评估快3～5倍。Traditional ET（ Expression Tree ） -based GEP has a major performance defect：Repetitive ET traverses and calculations. This paper proposes a novel model called Scale-Based GEP to solve the problem. Variable matrix is used to avoid repetitive calculations and traversing in ET-based gene evaluation. Experiments show that Scale-based evaluation outperforms ET-based method 3-5 times constantly.

关 键 词：数据挖掘 基因表达式编程 基于Scale的基因表达 

分 类 号：TP311.131[自动化与计算机技术—计算机软件与理论]