Journal of Computer Applications ›› 2015, Vol. 35 ›› Issue (11): 3218-3221.DOI: 10.11772/j.issn.1001-9081.2015.11.3218

• CRSSC 2015 Paper • Previous Articles     Next Articles

Multi-label learning with label-specific feature reduction

XU Suping, YANG Xibei, QI Yunsong   

  1. School of Computer Science and Engineering, Jiangsu University of Science and Technology, Zhenjiang Jiangsu 212003, China
  • Received:2015-06-20 Revised:2015-07-13 Published:2015-11-13


徐苏平, 杨习贝, 祁云嵩   

  1. 江苏科技大学 计算机科学与工程学院, 江苏 镇江 212003
  • 通讯作者: 杨习贝(1980-),男,江苏镇江人,副教授,博士,主要研究方向:粗糙集理论、粒计算、机器学习.
  • 作者简介:徐苏平(1991-),男,江苏扬州人,硕士研究生,主要研究方向:机器学习、粗糙集理论; 祁云嵩(1967-),男,江苏如皋人,教授,博士,主要研究方向:机器学习、智能信息处理.
  • 基金资助:

Abstract: In multi-label learning, since different labels may have their own characteristics, multi-label learning approach with label-specific features named LIFT has been proposed. However, the construction of label-specific features may increase the dimension of feature vector, which brings some redundant information in feature space. To solve this problem, a multi-label learning approach named FRS-LIFT was presented, which can implement label-specific feature reduction by fuzzy rough set. FRS-LIFT contains four steps: construction of label-specific features, reduction of feature dimensionality, training of classification models and prediction of unknown samples. The experimental results on 5 multi-label datasets show that, compared with LIFT, the proposed method can not only reduce the dimension of label-specific features, but also achieve satisfactory performances in 5 evaluation metrics.

Key words: feature reduction, fuzzy rough set, label-specific feature, multi-label learning

摘要: 在多标记学习中,由于不同的标记可能会带有自身的一些特性,所以目前已经出现了基于标记类属属性的多标记学习算法LIFT.然而,类属属性的构建可能会增加属性向量的维度,致使属性空间存在冗余信息.为此,借助模糊粗糙集提出了一种能够进行类属属性约简的多标记学习算法FRS-LIFT,其包含4个步骤:类属属性构建、属性维度约简、分类模型训练和未知样本预测.在5个多标记数据集上的实验结果表明,该算法与LIFT算法相比,不仅能够降低类属属性维数,而且在5种多标记评价指标上均具有较好的实验效果.

关键词: 属性约简, 模糊粗糙集, 类属属性, 多标记学习

CLC Number: