计算机应用研究
計算機應用研究
계산궤응용연구
APPLICATION RESEARCH OF COMPUTERS
2015年
5期
1387-1389,1398
,共4页
贺邓超%张宏军%郝文宁%张睿
賀鄧超%張宏軍%郝文寧%張睿
하산초%장굉군%학문저%장예
特征选择%Parzen 窗%条件互信息%特征离散度
特徵選擇%Parzen 窗%條件互信息%特徵離散度
특정선택%Parzen 창%조건호신식%특정리산도
feature selection%Parzen window%conditional mutual information%feature dispersion
为解决连续值特征条件互信息计算困难和对多值特征偏倚的问题,提出了一种基于 Parzen 窗条件互信息计算的特征选择方法。该方法通过 Parzen 窗估计出连续值特征的概率密度函数,进而方便准确地计算出条件互信息;同时在评价准则中引入特征离散度作为惩罚因子,克服了条件互信息计算对于多值特征的偏倚,实现了对连续型数据的特征选择。实验证明,该方法能够达到与现有方法相当甚至更好的效果,是一种有效的特征选择方法。
為解決連續值特徵條件互信息計算睏難和對多值特徵偏倚的問題,提齣瞭一種基于 Parzen 窗條件互信息計算的特徵選擇方法。該方法通過 Parzen 窗估計齣連續值特徵的概率密度函數,進而方便準確地計算齣條件互信息;同時在評價準則中引入特徵離散度作為懲罰因子,剋服瞭條件互信息計算對于多值特徵的偏倚,實現瞭對連續型數據的特徵選擇。實驗證明,該方法能夠達到與現有方法相噹甚至更好的效果,是一種有效的特徵選擇方法。
위해결련속치특정조건호신식계산곤난화대다치특정편의적문제,제출료일충기우 Parzen 창조건호신식계산적특정선택방법。해방법통과 Parzen 창고계출련속치특정적개솔밀도함수,진이방편준학지계산출조건호신식;동시재평개준칙중인입특정리산도작위징벌인자,극복료조건호신식계산대우다치특정적편의,실현료대련속형수거적특정선택。실험증명,해방법능구체도여현유방법상당심지경호적효과,시일충유효적특정선택방법。
In order to solve the problems of calculating the conditional mutual information of continuous variables and bias of multi-value features,this paper proposed a novel feature selection method.The method was based on computing conditional mutual information with Parzen window called PCMIFS,which adopted Parzen window to estimate the probability density func-tion and compute conditional mutual information of continuous feature.And introduced a penalty factor,feature dispersion,to overcome the bias of multi-value features.The experiment results show that comparing several existing method,PCMIFS can attain better or comparable performance,and is an effective feature selection method.