计算机应用研究2017，Vol.34Issue(4)：1036-1041,6.DOI:10.3969/j.issn.1001-3695.2017.04.018
一种面向标签排序数据集的特征选择方法
Feature selection method for label ranking dataset
摘要
Abstract
This paper proposed a feature selection method according to the characteristic of label ranking problem.First,it defined some new neighborhood information measures based on the neighborhood rough set which could measure the relevance,redundancy and interaction of different kinds of features directly.Then,it put forward a feature selection algorithm based on the neighborhood interaction weight factor.Experimental results show that the algorithm can effectively eliminate the irrelevant features and redundant features of label ranking dataset without reducing the ranking accuracies.关键词
标签排序/邻域粗糙集/特征选择Key words
label ranking/neighborhood rough set/feature selection分类
信息技术与安全科学引用本文复制引用
曾子林..一种面向标签排序数据集的特征选择方法[J].计算机应用研究,2017,34(4):1036-1041,6.基金项目
国家自然科学基金资助项目(11501281) （11501281）
