特征选择措施的稳定性能

IF 0.5 4区数学 Q4 STATISTICS & PROBABILITY

Theory of Probability and its Applications Pub Date : 2024-05-02 DOI:10.1137/s0040585x97t991726

A. V. Bulinski

{"title":"特征选择措施的稳定性能","authors":"A. V. Bulinski","doi":"10.1137/s0040585x97t991726","DOIUrl":null,"url":null,"abstract":"Theory of Probability &Its Applications, Volume 69, Issue 1, Page 25-34, May 2024. <br/> In this paper, we prove that the monotonicity property of the stability measure for the feature (factor) selection introduced by Nogueira, Sechidis, and Brown [J. Mach. Learn. Res., 18 (2018), pp. 1--54] may not hold. Another monotonicity property takes place. We also show the cases in which it is possible to compare by certain parameters the matrices describing the operation of algorithms for identifying relevant features.","PeriodicalId":51193,"journal":{"name":"Theory of Probability and its Applications","volume":null,"pages":null},"PeriodicalIF":0.5000,"publicationDate":"2024-05-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Stability Properties of Feature Selection Measures\",\"authors\":\"A. V. Bulinski\",\"doi\":\"10.1137/s0040585x97t991726\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Theory of Probability &Its Applications, Volume 69, Issue 1, Page 25-34, May 2024. <br/> In this paper, we prove that the monotonicity property of the stability measure for the feature (factor) selection introduced by Nogueira, Sechidis, and Brown [J. Mach. Learn. Res., 18 (2018), pp. 1--54] may not hold. Another monotonicity property takes place. We also show the cases in which it is possible to compare by certain parameters the matrices describing the operation of algorithms for identifying relevant features.\",\"PeriodicalId\":51193,\"journal\":{\"name\":\"Theory of Probability and its Applications\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":0.5000,\"publicationDate\":\"2024-05-02\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Theory of Probability and its Applications\",\"FirstCategoryId\":\"100\",\"ListUrlMain\":\"https://doi.org/10.1137/s0040585x97t991726\",\"RegionNum\":4,\"RegionCategory\":\"数学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q4\",\"JCRName\":\"STATISTICS & PROBABILITY\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Theory of Probability and its Applications","FirstCategoryId":"100","ListUrlMain":"https://doi.org/10.1137/s0040585x97t991726","RegionNum":4,"RegionCategory":"数学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q4","JCRName":"STATISTICS & PROBABILITY","Score":null,"Total":0}

引用次数: 0

摘要

概率论及其应用》（Theory of Probability &Its Applications），第 69 卷第 1 期，第 25-34 页，2024 年 5 月。在本文中，我们证明了 Nogueira、Sechidis 和 Brown [J. Mach. Learn. Res., 18 (2018), pp.另一个单调性属性发生了。我们还展示了可以通过某些参数比较描述识别相关特征的算法操作矩阵的情况。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

查看原文本刊更多论文

Stability Properties of Feature Selection Measures

Theory of Probability &Its Applications, Volume 69, Issue 1, Page 25-34, May 2024.
In this paper, we prove that the monotonicity property of the stability measure for the feature (factor) selection introduced by Nogueira, Sechidis, and Brown [J. Mach. Learn. Res., 18 (2018), pp. 1--54] may not hold. Another monotonicity property takes place. We also show the cases in which it is possible to compare by certain parameters the matrices describing the operation of algorithms for identifying relevant features.

求助全文

通过发布文献求助，成功后即可免费获取论文全文。去求助

来源期刊

Theory of Probability and its Applications 数学-统计学与概率论

CiteScore

1.00

自引率

16.70%

发文量

审稿时长

6 months

期刊介绍： Theory of Probability and Its Applications (TVP) accepts original articles and communications on the theory of probability, general problems of mathematical statistics, and applications of the theory of probability to natural science and technology. Articles of the latter type will be accepted only if the mathematical methods applied are essentially new.