模糊探勘程序來挖掘序列樣式; A Fuzzy Mining Process for Discovering Sequential Patterns

NCU Institutional Repository > 管理學院 > 資訊管理研究所 > 博碩士論文 > Item 987654321/13138

jsp.display-item.identifier=請使用永久網址來引用或連結此文件: http://ir.lib.ncu.edu.tw/handle/987654321/13138

题名:	模糊探勘程序來挖掘序列樣式;A Fuzzy Mining Process for Discovering Sequential Patterns
作者:	黃正魁;Cheng-Kui Huang
贡献者:	資訊管理研究所
关键词:	資料探勘;序列樣式;模糊集合;時間區間;多階層;數量資料;multi-level;time interval;fuzzy sets;sequential patterns;data mining;quantitative data
日期:	2006-05-25
上传时间:	2009-09-22 15:24:34 (UTC+8)
出版者:	國立中央大學圖書館
摘要:	隨著資料的大量增加，資料探勘(Data Mining)已經被使用在處理資料過剩的問題，並且在既有的資料中，去挖掘有用的、新的和具有潛力的樣式。然而，我們在挖掘量化型的資料(Quantitative Data)時，卻可能產生傳統不是0就是1的切割問題(Sharp Boundary Problem)，而這問題是傳統資料探勘方法無法解決的。為了這個問題，已經有許多學者，運用模糊集合(Fuzzy Sets)去挖掘帶有數量資料的樣式，尤其是在序列樣式(Sequential Patterns)的挖掘[18][25]。為了有更一般化的觀點來看資料探勘和模糊領域的結合，而幫助去挖掘序列樣式，本研究提出了一個模糊探勘的運作程序，來引導如何挖掘序列樣式(Fuzzy Mining Process for Discovering Sequential Patterns, FMPDSP)。此程序的目的是建立一個跨兩個領域合作的橋樑，進而瞭解並分析模糊序列樣式探勘的研究步驟。另外，本研究提出了三種不同的模糊序列樣式的研究，來證明這個新程序的可行性(Workable)和其一般化(Generalization)，並引導這兩個領域結合的新研究。 With the increase of data, data mining has been introduced to solve the overloading problem and to discover valid, novel, potentially useful patterns in existing data. In order to discover quantitative data, we may encounter a sharp boundary problem which the traditional data mining techniques cannot overcome. In view of this weakness, a lot of researches have been applied fuzzy sets to discover a variety of quantitative patterns, especially in sequential pattern mining [18][25]. Therefore, we devote to proposing a work process, Fuzzy Mining Process for Discovering Sequential Patterns (FMPDSP), to hold more general viewpoint combining Data Mining and Fuzzy Sets fields for discovering sequential patterns. The purpose of the process is to establish a cooperative relationship for the both fields to understand and analyze the investigating steps of fuzzy sequential pattern mining. Three researches were proposed to demonstrate that the FMPDSP can be workable and generalization to lead the future studies in the both fields.
显示于类别:	[資訊管理研究所] 博碩士論文

文件中的档案:

档案	大小	格式	浏览次数

在NCUIR中所有的数据项都受到原著作权保护.

社群 sharing

数据加载中.....