完整後設資料紀錄
DC 欄位語言
dc.contributor.authorLin, Yuh-Chi
dc.contributor.authorHadingham, Paul
dc.date.accessioned2009-08-23T04:40:22Z
dc.date.accessioned2020-05-25T06:25:48Z-
dc.date.available2009-08-23T04:40:22Z
dc.date.available2020-05-25T06:25:48Z-
dc.date.issued2006-10-20T19:03:33Z
dc.date.submitted1998-12-17
dc.identifier.urihttp://dspace.lib.fcu.edu.tw/handle/2377/2082-
dc.description.abstractFinding techniques for constructing a personalized web is an important issue in the World Wide Web research. In this paper, we explore the problem of dynamical personalized web promotion-how to effectively interpret dynamic user trends and how to efficiently promote the most relevant pages to users. First, we present a simple idea based on dynamic transactions to effectively represent the changing trends of web users over time. Second, we formalize the dynamical personalized web promotion problem as a similarity matching between a dynamic transaction and a group of clustered web logs. The dynamic promoting mechanism is designed for this purpose where two algorithms, cluster identification and page identification, are also presented. Finally, we introduce our method for efficiently clustering web introduce our method for efficiently clustering web logs. Parts of the ideas above have been implemented, and the preliminary results are encouraging.
dc.description.sponsorship成功大學,台南市
dc.format.extent7p.
dc.format.extent574863 bytes
dc.format.mimetypeapplication/pdf
dc.language.isozh_TW
dc.relation.ispartofseries1998 ICS會議
dc.subject.otherInternet Applications
dc.titleA New Agent-Oriented Approach for Personalized Web Management
分類:1998年 ICS 國際計算機會議

文件中的檔案:
檔案 描述 大小格式 
ce07ics001998000201.pdf561.39 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。