摘要
融合了基于数据点拟合的公式发现和因式分解技术,提出并实现了基于基因表达式编程(Gene Expression Programming,GEP)的多因子曲线拟合MFF(Multiple Factor Fitting)。利用MFF算法能够直接由客观数据挖掘出多个多项式乘积形式的函数关系公式以拟合原始数据集所表示的曲线。MFF中采用了有特色的概率相关系数对GEP中的适应度函数进行优化,使得精度提高了27%。同时采用阈值递减序列TDQ(Threshold Degression Queue)使得GEP成功率比传统技术提高了最大58倍。
This paper proposes an approach to implement function fitting by multiple factors named MFF（Muhiple Factor Fitting） based on GEP（Gene Expression Programming）.MFF can discover a function formed by multiple factors to fit the original curve. MFF optimizes the fitness function in GEP by special approach called probability correlation factor,which increases the precision by 27%.At the same time,adopting TDQ（Threshold Degression Queue） to improve the success-probability by 58 times compared with traditional approaches.
出处
《计算机工程与应用》
CSCD
北大核心
2007年第9期157-160,共4页
Computer Engineering and Applications
关键词
多因子曲线拟合
多项式分解
基因表达式编程
Multiple Factor Fitting
Polynomial Functions Factorization
Gene Expression Programming
