基于遗传算法的针灸处方寻优设计  

Design of Acupuncture Prescription Optimization Based on Genetic Algorithm

作　　者：高燕[1] 任玉兰[2] 周小平[1] 梁繁荣[2] 

机构地区：[1]成都信息工程学院,四川成都610225 [2]成都中医药大学,四川成都611137

出　　处：《成都信息工程学院学报》2012年第5期449-452,共4页Journal of Chengdu University of Information Technology

基　　金：国家重点基础研究发展计划(973计划)资助项目(2006CB504501);国家自然基金青年资助项目(81102742);四川省科技厅支撑资助项目(2007SGY008);成都信息工程学院科研基金(CRF201040)对本文的资助

摘　　要：几千年来针灸处方数据的积累,为现代医务工作者提供了丰富的针灸临床经验,但是,如何借鉴海量的针灸处方生成有价值、最优的针灸处方,则是现代针灸学致力于研究的重点。搜集了跨度几千年的大部分针灸处方数据,在对其进行结构化的基础上,利用遗传算法的思想对每一朝代单独聚类产生各自的适应度函数,然后再对各个朝代的适应度函数按其权重组合形成总的适应度函数,最终通过适应度函数取值的高低获得优化后的针灸处方。将算法用于开发针灸处方决策支持系统,经临床实践,其寻优结果具有较高的适用价值。For thousands of years of accumulation, acupuncture prescription data for the modern medical workers pro- vide rich clinical experience, but, how to generate the best acupuncture prescription according to mass of data, is dedicated to the research by science of acupuncture and moxibustion. This paper collects the most of the acupuncture prescription data spanning several thousand years, and on the basis of data structure, uses the genetic algorithm to construct their separate the fitness function according to every dynasty clustering, and then the fitness function of various dynasties according to their weight combined eventually general fitness function. With the general fitness function, the best acupuncture prescription is figured out. The algorithm is used to develop acupuncture prescription decision support system, the optimal result has high applicable value in the clinical practice.

关 键 词：计算机软件与理论 数据挖掘 遗传算法 针灸处方 基因编码 聚类算法 适应度函数 

分 类 号：TP311.135[自动化与计算机技术—计算机软件与理论]