基于块缩放的未标记数据集分类精度估计

IF 1.3 Q3 ENGINEERING, MULTIDISCIPLINARY
None Shingchern D. You, None Kai-Rong Lin, None Chien-Hung Liu
{"title":"基于块缩放的未标记数据集分类精度估计","authors":"None Shingchern D. You, None Kai-Rong Lin, None Chien-Hung Liu","doi":"10.46604/ijeti.2023.11975","DOIUrl":null,"url":null,"abstract":"This paper proposes an approach called block scaling quality (BSQ) for estimating the prediction accuracy of a deep network model. The basic operation perturbs the input spectrogram by multiplying all values within a block by , where is equal to 0 in the experiments. The ratio of perturbed spectrograms that have different prediction labels than the original spectrogram to the total number of perturbed spectrograms indicates how much of the spectrogram is crucial for the prediction. Thus, this ratio is inversely correlated with the accuracy of the dataset. The BSQ approach demonstrates satisfactory estimation accuracy in experiments when compared with various other approaches. When using only the Jamendo and FMA datasets, the estimation accuracy experiences an average error of 4.9% and 1.8%, respectively. Moreover, the BSQ approach holds advantages over some of the comparison counterparts. Overall, it presents a promising approach for estimating the accuracy of a deep network model.","PeriodicalId":43808,"journal":{"name":"International Journal of Engineering and Technology Innovation","volume":null,"pages":null},"PeriodicalIF":1.3000,"publicationDate":"2023-09-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Estimating Classification Accuracy for Unlabeled Datasets Based on Block Scaling\",\"authors\":\"None Shingchern D. You, None Kai-Rong Lin, None Chien-Hung Liu\",\"doi\":\"10.46604/ijeti.2023.11975\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"This paper proposes an approach called block scaling quality (BSQ) for estimating the prediction accuracy of a deep network model. The basic operation perturbs the input spectrogram by multiplying all values within a block by , where is equal to 0 in the experiments. The ratio of perturbed spectrograms that have different prediction labels than the original spectrogram to the total number of perturbed spectrograms indicates how much of the spectrogram is crucial for the prediction. Thus, this ratio is inversely correlated with the accuracy of the dataset. The BSQ approach demonstrates satisfactory estimation accuracy in experiments when compared with various other approaches. When using only the Jamendo and FMA datasets, the estimation accuracy experiences an average error of 4.9% and 1.8%, respectively. Moreover, the BSQ approach holds advantages over some of the comparison counterparts. Overall, it presents a promising approach for estimating the accuracy of a deep network model.\",\"PeriodicalId\":43808,\"journal\":{\"name\":\"International Journal of Engineering and Technology Innovation\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":1.3000,\"publicationDate\":\"2023-09-28\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"International Journal of Engineering and Technology Innovation\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.46604/ijeti.2023.11975\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q3\",\"JCRName\":\"ENGINEERING, MULTIDISCIPLINARY\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"International Journal of Engineering and Technology Innovation","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.46604/ijeti.2023.11975","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"ENGINEERING, MULTIDISCIPLINARY","Score":null,"Total":0}
引用次数: 0

摘要

本文提出了一种块缩放质量(BSQ)方法来估计深度网络模型的预测精度。基本操作通过将块内的所有值乘以,其中在实验中等于0,来扰动输入谱图。与原始谱图具有不同预测标签的摄动谱图与摄动谱图总数的比值表明了谱图中有多少对预测至关重要。因此,该比率与数据集的准确性呈负相关。实验结果表明,该方法具有较好的估计精度。当仅使用Jamendo和FMA数据集时,估计精度的平均误差分别为4.9%和1.8%。此外,BSQ方法比一些比较方法具有优势。总的来说,它为估计深度网络模型的精度提供了一种很有前途的方法。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
Estimating Classification Accuracy for Unlabeled Datasets Based on Block Scaling
This paper proposes an approach called block scaling quality (BSQ) for estimating the prediction accuracy of a deep network model. The basic operation perturbs the input spectrogram by multiplying all values within a block by , where is equal to 0 in the experiments. The ratio of perturbed spectrograms that have different prediction labels than the original spectrogram to the total number of perturbed spectrograms indicates how much of the spectrogram is crucial for the prediction. Thus, this ratio is inversely correlated with the accuracy of the dataset. The BSQ approach demonstrates satisfactory estimation accuracy in experiments when compared with various other approaches. When using only the Jamendo and FMA datasets, the estimation accuracy experiences an average error of 4.9% and 1.8%, respectively. Moreover, the BSQ approach holds advantages over some of the comparison counterparts. Overall, it presents a promising approach for estimating the accuracy of a deep network model.
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
CiteScore
2.80
自引率
0.00%
发文量
18
审稿时长
12 weeks
期刊介绍: The IJETI journal focus on the field of engineering and technology Innovation. And it publishes original papers including but not limited to the following fields: Automation Engineering Civil Engineering Control Engineering Electric Engineering Electronic Engineering Green Technology Information Engineering Mechanical Engineering Material Engineering Mechatronics and Robotics Engineering Nanotechnology Optic Engineering Sport Science and Technology Innovation Management Other Engineering and Technology Related Topics.
×
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