摘要
随着数据获取工具的快速发展,实际获取到的数据通常是动态更新的,且更新的速度也越来越快.数据集的动态更新通常有三中类型,即数据规模的更新、维数的更新和数据取值的更新.为此,本文针对含有缺失数据的数据集,通过讨论和分析信息熵随数据变化的更新机制,设计了一种基于信息熵的动态特征选择算法.算法中以信息熵的变化作为特征重要度的度量,进而给出了含有缺失数据的数据集中特征选择结果随数据取值变化的动态更新策略.实验结果进一步验证了新算法的可行性和高效性.
With the fast development of data acquisition tools,many real data always vary dynamically in much faster speed. Techniques mainly address data updating from three aspects：data expansion,increasing dimension and variation of data values. In this paper,by analyzing the updating mechanism of information entropy,a dynamic feature selection algorithm is proposed. On the basis of redefining the feature significance,a new updating strategy of feature selection is proposed for dealing with data set with dynamically varying data values. Experimental results show that the new algorithm is feasible and efficiency.
作者
王锋
宋鹏
WANG Feng;SONG Peng(School of Computer and Information Technology,Shanxi University,Taiyuan 030006,China;School of Economics and Management,Shanxi University,Taiyuan 030006,China)
出处
《小型微型计算机系统》
CSCD
北大核心
2018年第12期2585-2589,共5页
Journal of Chinese Computer Systems
基金
国家自然科学基金项目(61402272)资助
山西省教育厅高效科技创新项目(2016111)资助
关键词
缺失数据
动态数据集
特征选择
信息熵
data sets with missing data
dynamic data sets
feature selection
information entropy
