摘要
自20世纪90年代以来,特征选择成为模式识别和机器学习领域的重要研究方向,研究成果十分显著,但是也存在许多问题需要进一步研究。本文首先将特征选择视为特征集合空间中的启发式搜索问题,对特征选择涉及的四个要素进行了阐述,然后从各个角度对特征选择算法进行了分类,概述了其各个分支的发展态势,最后探讨了基于多目标免疫优化的特征选择方法的研究思路。
Feature selection has been an important research area in pattern recognition and machine learning since 90＇s of the 20th century. Great achievements have been achieved, however many problems remain to be unsolved and need further investigation. In this paper, we first describe feature selection in terms of heuristic search through the space of feature sets, discussing the four factors in feature selection algorithms,then classify many popular feature selection algorithms from different points of view and introduce several embranchments of feature selection and the development. At last, we discuss the research thought of a new feature selection algorithm based on multi-objective immune optimization method.
出处
《电子设计工程》
2011年第9期46-51,共6页
Electronic Design Engineering
基金
国家自然科学基金(60970082)
国家自然科学基金(50778109)
浙江省自然科学基金项目(Y1080777
Y3090061
Y3080457)
关键词
特征选择
特征子集
搜索
多目标优化
feature selection
features subset
search
multi-objective optimization
