Novel Apriori-Based Multi-Label Learning Algorithm by Exploiting Coupled Label Relationship  

作　　者：Zhenwu Wang Longbing Cao 

机构地区：[1]Department of Computer Science and Technology,China University of Mining and Technology(Beijing) [2]The Advanced Analytics Institute,University of Technology,Sydney

出　　处：《Journal of Beijing Institute of Technology》2017年第2期206-214,共9页北京理工大学学报（英文版）

基　　金：Supported by Australian Research Council Discovery(DP130102691);the National Science Foundation of China(61302157);China National 863 Project(2012AA12A308);China Pre-research Project of Nuclear Industry(FZ1402-08)

摘　　要：It is a key challenge to exploit the label coupling relationship in multi-label classification（MLC）problems.Most previous work focused on label pairwise relations,in which generally only global statistical information is used to analyze the coupled label relationship.In this work,firstly Bayesian and hypothesis testing methods are applied to predict the label set size of testing samples within their k nearest neighbor samples,which combines global and local statistical information,and then apriori algorithm is used to mine the label coupling relationship among multiple labels rather than pairwise labels,which can exploit the label coupling relations more accurately and comprehensively.The experimental results on text,biology and audio datasets shown that,compared with the state-of-the-art algorithm,the proposed algorithm can obtain better performance on 5 common criteria.It is a key challenge to exploit the label coupling relationship in multi-label classification（MLC）problems.Most previous work focused on label pairwise relations,in which generally only global statistical information is used to analyze the coupled label relationship.In this work,firstly Bayesian and hypothesis testing methods are applied to predict the label set size of testing samples within their k nearest neighbor samples,which combines global and local statistical information,and then apriori algorithm is used to mine the label coupling relationship among multiple labels rather than pairwise labels,which can exploit the label coupling relations more accurately and comprehensively.The experimental results on text,biology and audio datasets shown that,compared with the state-of-the-art algorithm,the proposed algorithm can obtain better performance on 5 common criteria.

关 键 词：multi-label classification hypothesis testing k nearest neighbor apriori algorithm label coupling 

分 类 号：TP181[自动化与计算机技术—控制理论与控制工程]