闽南师范大学学报(自然科学版)
閩南師範大學學報(自然科學版)
민남사범대학학보(자연과학판)
Journal of Zhangzhou Teachers College (Natural Science Edition)
2015年
3期
21-30
,共10页
粗糙集%动态%代价敏感%属性选择%误差范围%可变代价
粗糙集%動態%代價敏感%屬性選擇%誤差範圍%可變代價
조조집%동태%대개민감%속성선택%오차범위%가변대개
rough set%dynamic%cost sensitive%feature selection%error range%variable cost
代价是现实数据的重要方面。数据的测试代价与数据的误差范围,即数据的粒度紧密相关,而误分类代价又跟测试代价有关,已有的属性选择方法往往忽视了这一点。为了处理这种情况,提出了一种基于误差范围和可变代价的最优属性子集选择方法。首先建立了该方法的理论框架,再设计了相应算法。在该方法中,测试代价和误分类代价根据不同的误差置信水平自适应地生成。再以最小化平均总代价为目标进行属性选择,从而得到最优的属性子集和误差置信水平。实验结果验证了所提方法的有效性。
代價是現實數據的重要方麵。數據的測試代價與數據的誤差範圍,即數據的粒度緊密相關,而誤分類代價又跟測試代價有關,已有的屬性選擇方法往往忽視瞭這一點。為瞭處理這種情況,提齣瞭一種基于誤差範圍和可變代價的最優屬性子集選擇方法。首先建立瞭該方法的理論框架,再設計瞭相應算法。在該方法中,測試代價和誤分類代價根據不同的誤差置信水平自適應地生成。再以最小化平均總代價為目標進行屬性選擇,從而得到最優的屬性子集和誤差置信水平。實驗結果驗證瞭所提方法的有效性。
대개시현실수거적중요방면。수거적측시대개여수거적오차범위,즉수거적립도긴밀상관,이오분류대개우근측시대개유관,이유적속성선택방법왕왕홀시료저일점。위료처리저충정황,제출료일충기우오차범위화가변대개적최우속성자집선택방법。수선건립료해방법적이론광가,재설계료상응산법。재해방법중,측시대개화오분류대개근거불동적오차치신수평자괄응지생성。재이최소화평균총대개위목표진행속성선택,종이득도최우적속성자집화오차치신수평。실험결과험증료소제방법적유효성。
Cost is important to data in real application. Test costs of data are sensitive to error ranges, namely, the granularity of data, while misclassification costs are also related to test costs. The existing feature selection methods often ignore it. To address this situation, an approach is proposed for selecting optimal feature subset with error ranges and variable costs. Firstly, the theoretical framework is established. Then the corresponding algorithms are designed. In the method, test costs and misclassification costs are adaptively computed according to the confidence level of measurement errors. The objective of feature selection is to minimize the average total cost. By the method, the optimal feature subset and the best confidence level of errors can be obtained. The experimental results manifest the effectiveness of the proposed approach.