计算机科学
聚类分析
人工智能
机器学习
模棱两可
适应(眼睛)
多标签分类
数据挖掘
光学
物理
程序设计语言
出处
期刊:IEEE Transactions on Knowledge and Data Engineering
[Institute of Electrical and Electronics Engineers]
日期:2016-03-23
卷期号:28 (7): 1734-1748
被引量:543
标识
DOI:10.1109/tkde.2016.2545658
摘要
Although multi-label learning can deal with many problems with label ambiguity, it does not fit some real applications well where the overall distribution of the importance of the labels matters. This paper proposes a novel learning paradigm named label distribution learning (LDL) for such kind of applications. The label distribution covers a certain number of labels, representing the degree to which each label describes the instance. LDL is a more general learning framework which includes both single-label and multi-label learning as its special cases. This paper proposes six working LDL algorithms in three ways: problem transformation, algorithm adaptation, and specialized algorithm design. In order to compare the performance of the LDL algorithms, six representative and diverse evaluation measures are selected via a clustering analysis, and the first batch of label distribution datasets are collected and made publicly available. Experimental results on one artificial and 15 real-world datasets show clear advantages of the specialized algorithms, which indicates the importance of special design for the characteristics of the LDL problem.
科研通智能强力驱动
Strongly Powered by AbleSci AI