采购数据异常值检测方法探讨

K. Kono, Yoshiro Yamamoto
{"title":"采购数据异常值检测方法探讨","authors":"K. Kono, Yoshiro Yamamoto","doi":"10.1109/ICTKE.2016.7804092","DOIUrl":null,"url":null,"abstract":"This study essentially evaluates the purchasing histories of supermarkets. Supermarket purchasing-data is often dynamic and can significantly fluctuate via seasonal, bargain-sale, and regular-sale trends. When data are analyzed inclusive of such trends, it is usually necessary to account for these trends within the acquired statistical sets to prevent biasing of results. For this subject evaluation, time-series data were evaluated for each day during a two-year period; however, data trends were ultimately extensively modified during the subject collection period. In addition, detection of outliers (representing a higher day's sales)was successfully carried out at any given time during the data acquisition interval. In this analysis, a two-step process was ultimately employed that converted non-stationary data into stationary data. Firstly, a period of data was contracted; secondly, trend-removal of data was performed through component decomposition (via the use of a state-space model). As the study team applied these two approaches, outliers were regularly detected. Identification of a product sale date, however, could not be adequately represented within the estimated component of a state-space model.","PeriodicalId":321584,"journal":{"name":"2016 14th International Conference on ICT and Knowledge Engineering (ICT&KE)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2016-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Discussion of outlier detection methods of purchasing data\",\"authors\":\"K. Kono, Yoshiro Yamamoto\",\"doi\":\"10.1109/ICTKE.2016.7804092\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"This study essentially evaluates the purchasing histories of supermarkets. Supermarket purchasing-data is often dynamic and can significantly fluctuate via seasonal, bargain-sale, and regular-sale trends. When data are analyzed inclusive of such trends, it is usually necessary to account for these trends within the acquired statistical sets to prevent biasing of results. For this subject evaluation, time-series data were evaluated for each day during a two-year period; however, data trends were ultimately extensively modified during the subject collection period. In addition, detection of outliers (representing a higher day's sales)was successfully carried out at any given time during the data acquisition interval. In this analysis, a two-step process was ultimately employed that converted non-stationary data into stationary data. Firstly, a period of data was contracted; secondly, trend-removal of data was performed through component decomposition (via the use of a state-space model). As the study team applied these two approaches, outliers were regularly detected. Identification of a product sale date, however, could not be adequately represented within the estimated component of a state-space model.\",\"PeriodicalId\":321584,\"journal\":{\"name\":\"2016 14th International Conference on ICT and Knowledge Engineering (ICT&KE)\",\"volume\":\"1 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2016-11-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2016 14th International Conference on ICT and Knowledge Engineering (ICT&KE)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/ICTKE.2016.7804092\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2016 14th International Conference on ICT and Knowledge Engineering (ICT&KE)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICTKE.2016.7804092","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0

摘要

这项研究基本上评估了超市的购买历史。超市的采购数据通常是动态的,并且会随着季节、廉价销售和常规销售趋势而大幅波动。当分析包含这些趋势的数据时,通常有必要在获得的统计集中考虑这些趋势,以防止结果产生偏差。在这项受试者评估中,对两年期间每天的时间序列数据进行评估;然而,在主题收集期间,数据趋势最终被广泛修改。此外,在数据采集间隔期间的任何给定时间,都成功地进行了异常值(代表较高日销售额)的检测。在此分析中,最终采用两步过程将非平稳数据转换为平稳数据。首先,收缩一段时间的数据;其次,通过组件分解(通过使用状态空间模型)执行数据的趋势去除。当研究小组应用这两种方法时,会定期检测到异常值。但是,产品销售日期的标识不能在状态空间模型的估计组件中充分表示。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
Discussion of outlier detection methods of purchasing data
This study essentially evaluates the purchasing histories of supermarkets. Supermarket purchasing-data is often dynamic and can significantly fluctuate via seasonal, bargain-sale, and regular-sale trends. When data are analyzed inclusive of such trends, it is usually necessary to account for these trends within the acquired statistical sets to prevent biasing of results. For this subject evaluation, time-series data were evaluated for each day during a two-year period; however, data trends were ultimately extensively modified during the subject collection period. In addition, detection of outliers (representing a higher day's sales)was successfully carried out at any given time during the data acquisition interval. In this analysis, a two-step process was ultimately employed that converted non-stationary data into stationary data. Firstly, a period of data was contracted; secondly, trend-removal of data was performed through component decomposition (via the use of a state-space model). As the study team applied these two approaches, outliers were regularly detected. Identification of a product sale date, however, could not be adequately represented within the estimated component of a state-space 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
自引率
0.00%
发文量
0
×
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:604180095
Book学术官方微信