Multi-label Learning of Kernel Extreme Learning Machine with Non-Equilibrium Label Completion

doi:10.3969/j.issn.0372-2112.2019.03.029

您当前的位置：

首页 >

文章列表页 >

Multi-label Learning of Kernel Extreme Learning Machine with Non-Equilibrium Label Completion

更新时间：2025-07-02

- Multi-label Learning of Kernel Extreme Learning Machine with Non-Equilibrium Label Completion
- Acta Electronica Sinica Vol. 47, Issue 3, Pages: 719-725(2019)
- 作者机构：
  
  1. 安庆师范大学计算机与信息学院,安徽,安庆,246011
  2. 安徽省高校智能感知与计算重点实验室,安徽,安庆,246011
  3. 安庆师范大学计算机与信息学院,安徽,安庆,246011
  4. 安徽省高校智能感知与计算重点实验室,安徽,安庆,246011
- 作者简介：
- 基金信息：
  
  Key Project of Natural Science Foundation in Universities of Anhui Province (No.KJ2017A352);Fund of Key Laboratory of Colleges and Universities in Anhui Province (No.ACAIM160102)
- DOI：10.3969/j.issn.0372-2112.2019.03.029
  CLC： TP18
- Published Online：25 March 2019，
  
  Published：2019
- 稿件说明：
移动端阅览
Multi-label Learning of Kernel Extreme Learning Machine with Non-Equilibrium Label Completion[J]. Acta Electronica Sinica, 2019, 47(3): 719-725.
DOI：

Multi-label Learning of Kernel Extreme Learning Machine with Non-Equilibrium Label Completion[J]. Acta Electronica Sinica, 2019, 47(3): 719-725. DOI： 10.3969/j.issn.0372-2112.2019.03.029.

摘要

目前众多的研究者通常直接将标签置信度矩阵作为先验知识直接加入到分类模型中，并没有考虑未标注先验知识对标签集质量的影响.基于此，引入非平衡参数的方法，将先验知识获得的基础置信度矩阵进行非平衡化，从而提出一种非平衡化的标签补全的核极限学习机多标签学习算法（KELM-NeLC）：首先使用信息熵计算标签之间的相关关系得到标签置信度矩阵，然后利用非平衡参数方法对基础的标签置信度矩阵进行改进，构建出一个非平衡的标签补全矩阵，最后为了学习获得更加准确的标签置信度矩阵，将非平衡化的标签补全矩阵与核极限学习机进行联合学习，依此解决多标签分类问题.提出的算法在公开的多个基准多标签数据集中的实验结果表明，KELM-NeLC算法较其他对比的多标签学习算法有一定优势，使用统计假设检验进一步说明所提出算法的有效性.

Abstract

At present

many researchers usually directly add the label confidence matrix as a priori knowledge to the classification model

and do not consider the influence of non-equilibrium prior knowledge on the quality of the label set.Based on this

the method of non-equilibrium parameters is introduced

and the basis confidence matrix obtained from the prior knowledge is non-equilibrium.Therefore

a multi-label learning algorithm is proposed

which uses kernel extreme learning machine with non-equilibrium label completion (KELM-NeLC).Firstly

information entropy is used to measure the correlation between labels which gets the basic label confidence matrix.Secondly

the basic label confidence matrix is improved to construct non-equilibrium label completion matrix by the non-equilibrium parameter.Finally

in order to learn to obtain a more accurate label confidence matrix

the non-equilibrium label completion matrix is introduced with the kernel extreme learning machine to solve the multi-label classification problem.The experimental results of the proposed algorithm in the opening benchmark multi-label datasets show that the KELM-NeLC algorithm has some advantages over other comparative multi-label learning algorithms and the statistical hypothesis test further illustrates the effectiveness of the proposed algorithm.

关键词

Keywords

references

Views

487

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Multi-Label Evolutionary Hypernetwork Based on Label Correlations

A Detection Method for Android Repackaged Malware Based on Sensitive Component Function Call Graph

A Method for Continuous Detection and Classification of Malicious Network Traffic Based on Double-Layer Model and Distribution of Indexes

Research on High-Efficiency Integrated Circuit DFT Technology Based on Machine Learning

DRHA-UIE: An Underwater Image Enhancement Method Based on Dual Residual Hybrid Attention Block

Related Author

WANG Jin

LIU Bin

SUN Kai-wei

CHEN Qiao-song

DENG Xin

DU Rui-ying

CHEN Jing

WU Cong

Related Institution

Chongqing Key Laboratory of Computational Intelligence, Chongqing University of Posts and Telecommunications

School of Computer and Information Engineering, Henan Normal University

School of Cyber Science and Engineering, Wuhan University

Key Laboratory of Aerospace Information Security and Trusted Computing, Ministry of Education, Wuhan University

Rizhao Institute of Information Technology, Wuhan University

⁰