[1]刘凯,郑山红,蒋权,等.基于随机森林的自适应特征选择算法[J].计算机技术与发展,2018,28(09):101-104.[doi:10．3969/j．issn．1673－629X．2018．09．021]
　LIU Kai,ZHENG Shanhong,JIANG Quan,et al.A Self－adaptive Feature Selection Algorithm Based on Ｒandom Forest[J].,2018,28(09):101-104.[doi:10．3969/j．issn．1673－629X．2018．09．021]
点击复制

基于随机森林的自适应特征选择算法()
分享到：

《计算机技术与发展》[ISSN:1006-6977/CN:61-1281/TN]

卷:
28
期数:
2018年09期
页码:
101-104
栏目:
智能、算法、系统工程
出版日期:
2018-09-10

文章信息/Info

Title:
A Self－adaptive Feature Selection Algorithm Based on Ｒandom Forest
文章编号:
1673－629X(2018)09－0101－04
作者:
刘凯; 郑山红; 蒋权; 赵天傲
长春工业大学 计算机科学与工程学院，吉林 长春 130012
Author(s):
LIU Kai; ZHENG Shan－hong; JIANG Quan; ZHAO Tian－ao
School of Computer Science and Engineering，Changchun University of Technology，Changchun 130012，China
关键词:
随机森林; 自适应; 特征选择; Group LASSO 方法
Keywords:
random forest; self－adaptive; feature selection; Group LASSO method
分类号:
TP301.6
DOI:
10．3969/j．issn．1673－629X．2018．09．021
文献标志码:
A
摘要:
为了解决传统的随机森林算法在随机特征选择时，导致少数比较重要的特征变量被过滤掉的问题，以及没有考虑特征变量相关性对预测应变量准确性带来的影响，提出了一种基于随机森林的自适应特征选择算法 SAＲFFS。该算法首先利用卡方检验样本间关联程度后自助采样，并设计出一种特征对类代表强弱程度的计算方法;然后引入自适应稀疏约束机制 Group LASSO 优化特征的选择;最后在 Spark 分布式计算平台利用 UCI 数据集进行实验，结果表明，相比传统的 RF算法，SARFFS 算法在特征子集选择上具有更好的性能，在 F 1 上提升将近 9%。从最终排名靠前的重要特征分析，该算法能够考虑特征间相关性，对预测结果确实有影响，并有效地提高了随机属性权值的可靠性和稳定性。
Abstract:
In order to solve the problem that a small number of important variables are filtered out in the selection process of random feature adopted by the method of traditional random forest algorithm，and without considering the influence of characteristic variable correlation on the accuracy of prediction variables，we propose an self－adaptive feature selection algorithm SAＲFFS based on random forests． It first uses the Chi square to test the degree of association between samples and then bootstrap sampling，and we design a method for calculating the intensity and degree of the class represented by the feature． Then，an adaptive sparse constraint mechanism Group LASSO is introduced to optimize the selection of the features． Finally，the experiments are carried out on the Spark distributed computing platform using UCI data sets，which shows that compared with the traditional ＲF algorithm，the SAＲFFS has better performance in feature subset selection，and the efficiency has been increased nearly 9% in the F 1 ． From the analysis of important characteristics of the final ranking，the proposed algorithm can consider the correlation having an effect on the prediction results definitely，and improves the reliability and stability of random attribute weights effectively．

相似文献/References:

[1]宋淑娜 李金霞 胡学坤 高尚.一种自适应模糊阈值区间的图像分割方法[J].计算机技术与发展,2010,(05):121.
　SONG Shu-na,LI Jin-xia,HU Xue-kun,et al.A Method of Adaptive Fuzzy Threshold Region for Image Segmentation[J].,2010,(09):121.
[2]董明忠.一种UWB Ad Hoc网络的自适应MAC协议算法与仿真[J].计算机技术与发展,2009,(08):92.
　DONG Ming-zhong.An Adaptive MAC Protocol Algorithm and Simulation Based on UWB Ad Hoc Networks[J].,2009,(09):92.
[3]王树梅 王志成 蔡健.一种基于灰度形态学的小波域边缘检测算法[J].计算机技术与发展,2009,(01):32.
　WANG Shu-mei,WANG Zhi-cheng,CAI Jian.A Novel Edge- Detection Algorithm in Wavelet Gray - Scale Morphology[J].,2009,(09):32.
[4]陈成 杨晨晖 聂文 龚元浩.基于浮游植物图像的模糊算子边缘检测算法[J].计算机技术与发展,2009,(03):22.
　CHEN Cheng,YANG Chen-hui,NIE Wen,et al.Based on Marine Phytoplankton Cells Images of Fuzzy Operator Edge Detection Algorithm[J].,2009,(09):22.
[5]邓秀勤 熊勇.用于图像处理的加权中值滤波算法[J].计算机技术与发展,2009,(03):46.
　DENG Xiu-qin,XIONG Yong.Weighted Median Filter Algorithm for Image Processing[J].,2009,(09):46.
[6]周俊明 胡小龙 彭建伟.功塞监控图形系统中自适应着色处理[J].计算机技术与发展,2008,(04):245.
　ZHOU Jun-ming,HU Xiao-long,PENG Jian-wei.Power-Aware Adaptive Shading for Graphics System[J].,2008,(09):245.
[7]赵纪涛 马莉 王现君 尚光龙.一种自适应的模糊关联规则挖掘算法[J].计算机技术与发展,2008,(05):64.
　ZHAO Ji- tao,MA Li,WANG Xian-jun,et al.An Adaptive Algorithm for Mining Fuzzy Association Rules[J].,2008,(09):64.
[8]陈珂 徐科[].全自动酶免工作站计算机控制系统设计[J].计算机技术与发展,2008,(06):160.
　CHEN Ke,XU Ke.Design of Computer Control System for Automated ELISA Workstation[J].,2008,(09):160.
[9]鲁群 周爱武.双变异算子遗传算法的应用[J].计算机技术与发展,2008,(07):42.
　LU Qun,ZHOU Ai-wu.Application of Genetic Algorithm Based on Dual Mutation[J].,2008,(09):42.
[10]张萍 刘弘.改进的IGA在建筑造型创新设计中的应用[J].计算机技术与发展,2008,(07):250.
　ZHANG Ping,LIU Hong.Improved IGA and Its Application in Construction Creative Design[J].,2008,(09):250.

更新日期/Last Update: 2018-09-10