Unsupervised Feature Selection with Latent Relationship Penalty Term

IF 1.9 3区 数学 Q1 MATHEMATICS, APPLIED
Axioms Pub Date : 2023-12-21 DOI:10.3390/axioms13010006
Ziping Ma, Yulei Huang, Huirong Li, Jingyu Wang
{"title":"Unsupervised Feature Selection with Latent Relationship Penalty Term","authors":"Ziping Ma, Yulei Huang, Huirong Li, Jingyu Wang","doi":"10.3390/axioms13010006","DOIUrl":null,"url":null,"abstract":"With the exponential growth of high dimensional unlabeled data, unsupervised feature selection (UFS) has attracted considerable attention due to its excellent performance in machine learning. Existing UFS methods implicitly assigned the same attribute score to each sample, which disregarded the distinctiveness of features and weakened the clustering performance of UFS methods to some extent. To alleviate these issues, a novel UFS method is proposed, named unsupervised feature selection with latent relationship penalty term (LRPFS). Firstly, latent learning is innovatively designed by assigning explicitly an attribute score to each sample according to its unique importance in clustering results. With this strategy, the inevitable noise interference can be removed effectively while retaining the intrinsic structure of data samples. Secondly, an appropriate sparse model is incorporated into the penalty term to further optimize its roles as follows: (1) It imposes potential constraints on the feature matrix to guarantee the uniqueness of the solution. (2) The interconnection between data instances is established by a pairwise relationship situation. Extensive experiments on benchmark datasets demonstrate that the proposed method is superior to relevant state-of-the-art algorithms with an average improvement of 10.17% in terms of accuracy.","PeriodicalId":53148,"journal":{"name":"Axioms","volume":"68 22","pages":""},"PeriodicalIF":1.9000,"publicationDate":"2023-12-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Axioms","FirstCategoryId":"100","ListUrlMain":"https://doi.org/10.3390/axioms13010006","RegionNum":3,"RegionCategory":"数学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"MATHEMATICS, APPLIED","Score":null,"Total":0}
引用次数: 0

Abstract

With the exponential growth of high dimensional unlabeled data, unsupervised feature selection (UFS) has attracted considerable attention due to its excellent performance in machine learning. Existing UFS methods implicitly assigned the same attribute score to each sample, which disregarded the distinctiveness of features and weakened the clustering performance of UFS methods to some extent. To alleviate these issues, a novel UFS method is proposed, named unsupervised feature selection with latent relationship penalty term (LRPFS). Firstly, latent learning is innovatively designed by assigning explicitly an attribute score to each sample according to its unique importance in clustering results. With this strategy, the inevitable noise interference can be removed effectively while retaining the intrinsic structure of data samples. Secondly, an appropriate sparse model is incorporated into the penalty term to further optimize its roles as follows: (1) It imposes potential constraints on the feature matrix to guarantee the uniqueness of the solution. (2) The interconnection between data instances is established by a pairwise relationship situation. Extensive experiments on benchmark datasets demonstrate that the proposed method is superior to relevant state-of-the-art algorithms with an average improvement of 10.17% in terms of accuracy.
带有潜在关系惩罚项的无监督特征选择
随着高维无标注数据的指数级增长,无监督特征选择(UFS)因其在机器学习中的卓越表现而备受关注。现有的无监督特征选择方法隐含地为每个样本分配相同的属性得分,这就忽略了特征的独特性,在一定程度上削弱了无监督特征选择方法的聚类性能。为了解决这些问题,我们提出了一种新型的 UFS 方法,即带有潜在关系惩罚项的无监督特征选择(LRPFS)。首先,创新性地设计了潜在学习方法,即根据每个样本在聚类结果中的独特重要性,为其分配明确的属性分数。通过这种策略,可以有效地消除不可避免的噪声干扰,同时保留数据样本的内在结构。其次,在惩罚项中加入适当的稀疏模型,以进一步优化其作用,具体如下:(1) 对特征矩阵施加潜在约束,保证解的唯一性。(2) 通过成对关系建立数据实例之间的相互联系。在基准数据集上进行的大量实验表明,所提出的方法优于相关的最先进算法,平均准确率提高了 10.17%。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 求助全文
来源期刊
Axioms
Axioms Mathematics-Algebra and Number Theory
自引率
10.00%
发文量
604
审稿时长
11 weeks
期刊介绍: Axiomatic theories in physics and in mathematics (for example, axiomatic theory of thermodynamics, and also either the axiomatic classical set theory or the axiomatic fuzzy set theory) Axiomatization, axiomatic methods, theorems, mathematical proofs Algebraic structures, field theory, group theory, topology, vector spaces Mathematical analysis Mathematical physics Mathematical logic, and non-classical logics, such as fuzzy logic, modal logic, non-monotonic logic. etc. Classical and fuzzy set theories Number theory Systems theory Classical measures, fuzzy measures, representation theory, and probability theory Graph theory Information theory Entropy Symmetry Differential equations and dynamical systems Relativity and quantum theories Mathematical chemistry Automata theory Mathematical problems of artificial intelligence Complex networks from a mathematical viewpoint Reasoning under uncertainty Interdisciplinary applications of mathematical theory.
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信