面向标签可变性的可拓分类方法

田英杰; 刘大莲; 李兴森

doi:10.12052/gdutxb.250108

欢迎来到能源电力期刊网！

English Version 期刊登录

您当前的位置：

首页 >

文章列表页 >

面向标签可变性的可拓分类方法

可拓学与创新方法 | 更新时间：2026-05-27

- 面向标签可变性的可拓分类方法
- Extension Classification Method for Label Variability
- 广东工业大学学报 2025年42卷第4期页码：1-7
- 作者机构：
  
  1. 中国科学院虚拟经济与数据科学研究中心,北京,100190
  2. 北京联合大学数理部,北京,100101
  3. 北京联合大学数理与交叉科学研究院,北京,100101
  4. 广东工业大学可拓学与创新方法研究所,广东,广州,510006
- 作者简介：
  
  田英杰 (1973–)，男，教授，博士，主要研究方向为机器学习与最优化，E-mail：tyj@ucas.ac.cn
  刘大莲 (1978–)，女，教授，博士，主要研究方向为机器学习与最优化，E-mail：ldtdalian@buu.edu.cn
- 基金信息：
  
  国家自然科学基金资助项目 (71731009，72071049);广东省自然科学基金资助项目(2024A1515011324);北京联合大学校级科研项目(ZK20202507)
- DOI：10.12052/gdutxb.250108
  中图分类号： TP181
- 纸质出版：2025
- 稿件说明：
移动端阅览
田英杰, 刘大莲, 李兴森. 面向标签可变性的可拓分类方法[J]. 广东工业大学学报, 2025, 42(4): 1–7.

Tian Yingjie, Liu Dalian, Li Xingsen. Extension Classification Method for Label Variability[J]. 2025, 42(4): 1-7.
田英杰, 刘大莲, 李兴森. 面向标签可变性的可拓分类方法[J]. 广东工业大学学报, 2025, 42(4): 1–7. DOI： 10.12052/gdutxb.250108

Tian Yingjie, Liu Dalian, Li Xingsen. Extension Classification Method for Label Variability[J]. 2025, 42(4): 1-7. DOI： 10.12052/gdutxb.250108.

摘要

传统分类方法通常假设训练样本的类别标签是静态且确定的，忽略了现实场景中样本标签可能随条件变化而发生转变的动态性特征。针对这一问题，本文提出了一种新的学习问题设定——标签可变性的可拓分类问题，在训练数据中同时标注样本的类别标签与标签可变性状态，用以刻画样本在变化机制作用下的类别转变潜力。基于此设定，设计了多标签学习框架，构建了基于支持向量机的可变标签可拓分类算法，实现类别判别与标签可变性预测的协同优化。实验部分在合成数据集与真实数据集上验证了所提算法的有效性。本文工作为复杂系统中标签动态性学习问题提供了一种新的建模思路，具有较好的应用前景。

Abstract

Traditional classification algorithms typically assume that the labels of training samples are static and deterministic

ignoring the dynamic characteristics of sample labels that may change with conditions in real-world scenarios. In response to this issue

this paper proposes a new learning problem setting—the Extended Classification Problem

which simultaneously gives the class labels and label variability states of samples in the training data

which to characterize the class transition potential of samples under the influence of change mechanisms. Based on this setting

a multi-label learning framework was designed

an extension classification algorithm for label variability using support vector machine was constructed

to achieve collaborative optimization of category discrimination and label variability prediction. The experimental section validated the effectiveness of the proposed algorithm on both synthetic and real datasets. This paper provides a new modeling approach for label dynamic learning problems

which has good application prospects.

关键词

Keywords

references

YANG C Y, LI X S. Development and research progress of extenics over the past 40 years[J]. Journal of Guangdong University of Technology, 2023, 40(6): 1-11.

LI X S, LU L, XU L B. Research on knowledge graph model of shili based on comparative analysis of WSR methodology and extenics[J]. Business Review, 2021, 33(5): 152-162.

LI X S, ZHANG H L, ZHU Z X, et al . An intelligent transformation knowledge mining method based on extenics[J ] . Journal of Internet Technology, 2013, 14(2): 315-325.

杨春燕, 李小妹, 陈文伟, 等. 可拓数据挖掘方法及其计算机实现[M]. 广州: 广东高等教育出版社, 2010

蔡文, 杨春燕, 陈文伟, 等.可拓集与可拓数据挖掘[M]. 北京: 科学出版社, 2008.

GUO C, PLEISS G, SUN Y, et al . On calibration of modern neural networks[C ] // arXiv: 1706.04599(2017-06-14)[2025-06-10 ] . https://doi.org/10.48550/arXiv.1706.04599.

BENGIO Y, LOURADOUR J, COLLOBERT R, et al . Curriculum learning[C ] //Proceedings of the 26th Annual International Conference on Machine Learning. Montreal Quebec: ICML, 2009: 41-48.

KUMAR M P, PACKER B, KOLLER D. Self-paced learning for latent variable models[C] //Advances in Neural Information Processing Systems. Vancouver: NeurIPS, 2010.

TONEVA M, SORDONI A, TACHET DES COMBES R, et al . An empirical study of example forgetting during deep neural network learning[C ] //International Conference on Learning Representations. New Orleans: ICLR, 2019.

GOODFELLOW I J, SHLENS J, SZEGEDY C. Explaining and harnessing adversarial examples[C] //International Conference on Learning Representations. San Diego: ICLR, 2015.

MADRY A, MAKELOV A, SCHMIDT L, et al . Towards deep learning models resistant to adversarial attacks[C ] //International Conference on Learning Representations. Vancouver: ICLR, 2018.

RIBEIRO M T, SINGH S, GUESTRIN C. “Why should I trust you?”: Explaining the predictions of any classifier[C] //Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. San Francisco: KDD, 2016: 1135-1144.

LUNDBERG S M, LEE S I. A unified approach to interpreting model predictions[C] //Advances in Neural Information Processing Systems. LongBeach: NeurIPS, 2017.

WACHTER S, MITTELSTADT B, RUSSELL C. Counterfactual explanations without opening the black box: Automated decisions and the GDPR[J]. Harvard Journal of Law Technology, 2017, 31: 841.

MOTHILAL R K, SHARMA A, TAN C. Explaining machine learning classifiers through diverse counterfactual explanations[C] //Proceedings of the 2020 ACM Conference on Fairness, Accountability, and Transparency. Online: FAccT 2020, 2020: 607-617.

ZHANG M L, ZHOU Z H. A review on multi-label learning algorithms[J]. IEEE Transactions on Knowledge and Data Engineering, 2014, 26(8): 1819-1837.

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

基于可拓学的多目标开放性问题求解方法研究

异类多维复合元间的传导规则及其在可拓创新设计中的应用

转换桥方法的改进及其应用

可拓学40年发展历程及研究进展

基于分合链方法的图的意大利支配数研究