完整後設資料紀錄
DC 欄位語言
dc.contributor.authorHarangsri, Banchong
dc.contributor.authorShepherd, John
dc.contributor.authorNgu, Anne
dc.date.accessioned2009-08-23T04:39:01Z
dc.date.accessioned2020-05-25T06:28:06Z-
dc.date.available2009-08-23T04:39:01Z
dc.date.available2020-05-25T06:28:06Z-
dc.date.issued2006-10-24T06:52:20Z
dc.date.submitted1996-12-19
dc.identifier.urihttp://dspace.lib.fcu.edu.tw/handle/2377/2390-
dc.description.abstractWe propose two novel notions in this paper:the first is that machine learning techniques can be used to solve the problem of query size estimation and the second is a new generic algorithm to correct the training set of queries in response to updates. The main advantage for machine learning is that no database scan is required to collect statistics for query size estimation. The training set correction algorithms is useful in that it allows us to “re-vitalise” some existing query size estimation methods whose performance previously deteriorated in the presence of high update loads. A by-product of his is that the length of training sets can be fixed – the size of the training set determines the level of error in query estimation. Our experimental results show the (1) the machine learning technique is superior to a recent curve fitting method in approximating query result sizes and (2) the machine learning technique still performs as well after the correction algorithm is applied.
dc.description.sponsorship中山大學,高雄市
dc.format.extent8p.
dc.format.extent612920 bytes
dc.format.mimetypeapplication/pdf
dc.language.isozh_TW
dc.relation.ispartofseries1996 ICS會議
dc.subjectQuery Size Estimation
dc.subjectQuery Optimisation
dc.subjectMachine Learning
dc.subject.otherMachine Learning
dc.titleQuery Size Estimation using Machine Learning
分類:1996年 ICS 國際計算機會議

文件中的檔案:
檔案 描述 大小格式 
ce07ics001996000063.pdf598.55 kBAdobe PDF檢視/開啟


在 DSpace 系統中的文件,除了特別指名其著作權條款之外,均受到著作權保護,並且保留所有的權利。