摘要
机器学习研究的是通过经验自动改进的计算机算法。本文提出对FICSEM的一种改进方法:FICSEM2。FICSEM是一种单例学习方法。适合大数据集的运算,是基于实验、开放的方法。FICSEM2对FICSEM中规则提取算法进行了改进,在计算机免疫的应用中验证FICSEM2的提取规则质量。
Machine Learning is the study of computer algorithms that improve automatically through experience. This paper puts for- ward FICSEM2 which improvement for FICSEM.FICSEM is an one-case study method. FICSEM is an open method, in which you can change the way of classing or study. It is based on experiments, and more fitted in large scale data sets. FICSEM2 is improve the way of building the rule sets for FICSEM. Application of computer immune to evaluate quality of the rules sets which are built by FICSEM2.
出处
《微计算机信息》
北大核心
2007年第30期193-194,312,共3页
Control & Automation
基金
河北省科学技术研究与发展计划资助(04213534
3213502)
关键词
复杂适应性系统
计算机免疫
单例学习
规则评价
Complex adaptive system(CAS), computer immune system, one-case study, evaluating of rules