MIMLRBF: RBF neural networks for multi-instance multi-label learning

作     者：Zhang, Min-Ling Wang, Zhi-Jian 

作者机构：Hohai Univ Coll Comp & Informat Engn Nanjing 210098 Peoples R China 

出 版 物：《NEUROCOMPUTING》 (神经计算)

年 卷 期：2009年第72卷第16-18期

页      面：3951-3956页

核心收录：

学科分类：08[工学] 0812[工学-计算机科学与技术（可授工学、理学学位）] 

基　　金：National Science Foundation of China Programs Foundation of Ministry of Education of China Open Foundation of National Key Laboratory for Novel Software Technology of China [KFKT2008BI2] Startup Foundation for Excellent New Faculties of Hohai University 

主　　题：Machine learning Multi-instance multi-label learning Radial basis function Scene classification Text categorization 

摘      要：In multi-instance multi-label learning (MIML), each example is not only represented by multiple instances but also associated with multiple class labels. Several learning frameworks, such as the traditional supervised learning. can be regarded as degenerated versions of MIML. Therefore, an intuitive way to solve MIML problem is to identify its equivalence in its degenerated versions. However, this identification process would make useful information encoded in training examples get lost and thus impair the learning algorithm s performance. In this paper, RBF neural networks are adapted to learn from MIML examples. Connections between instances and labels are directly exploited in the process of first layer clustering and second layer optimization. The proposed method demonstrates superior performance on two real-world MIML tasks. (c) 2009 Elsevier B.V. All rights reserved.