應用屬性值切割與基因分群技術以推估遺漏值

NHUIR > College of Science and Technology > Department of Information Management > Disserations and Theses > Item 987654321/20007

Please use this identifier to cite or link to this item: http://nhuir.nhu.edu.tw/handle/987654321/20007

Title:	應用屬性值切割與基因分群技術以推估遺漏值
Other Titles:	Applying Attribute Values Partitioning and GA Clustering Technique for Estimating Missing Values
Authors:	吳文盛 Wu, Wen-sheng
Contributors:	資訊管理學系碩士班邱宏彬 Hung-pin Chiu
Keywords:	群集分析;遺漏值;資料探勘;屬性值切割法;基因演算法 attribute values partitioning;clustering analysis;missing value estimation;data mining;genetic clustering algorithms
Date:	2009
Issue Date:	2015-03-25 15:25:15 (UTC+8)
Abstract:	資料探勘是由大量資料中挖掘出隠藏知識的重要技術，目前企業或政府各方面決策幾乎是以歷史資料探勘結果分析為基礎，故資料庫的完整性則十分的重要。若是資料庫中出現過多的遺漏值，則容易影響資料分析結果的有效性。我們以群集分析為基礎來建立一個遺漏值推估模組，將物以類聚、群內同質、群間異質的特性應用在遺漏值推估上。再利用屬性值切割法來找出屬性之間的關聯，讓分群後的資料關聯與特性更為緊密；另外基因演算法具備隨機多點搜尋與演化過程的特性，可經由不斷演化找出較佳的分群結果。所以本研究嘗試結合屬性值切割法與基因分群技術，來進行遺漏值的推估，讓使用者可以在使用資料探勘方法時仍可保有最大的資訊量，期望探勘出的結果更具意義。本研究將此推估模式應用到四個真實資料集上，以驗證本研究方法之可行性與推估效能。　　Data mining is a vitally important technique to uncover hidden information from a set of raw data. The managers can exploit the mining results to make effective decisions. However, missing data significantly distort data mining results. Therefore, data preprocessing of missing values is very critical in successful data mining. Data clustering techniques is the partitioning of a dataset into subsets so that the data in each subset share common pattern. The shared pattern can be utilized to estimate the missing values. In this study, we propose an attribute values partitioning technique to preserve the relationships between attributes for estimating missing values. In addition, genetic algorithm is a powerful population-based stochastic search process for finding the robust clustering result. Therefore, we also propose a genetic clustering-based approach to estimate the missing data. Furthermore, we integrate the attribute values partitioning with the genetic clustering techniques to improve the estimation performance. Effectiveness of the proposed approaches is demonstrated on four datasets for four different rates of missing data. The empirical evaluation shows the integrated missing data processing approach provides competitive results or performs well compared with the existing methods.
Appears in Collections:	[Department of Information Management] Disserations and Theses

Files in This Item:

File	Description	Size	Format
097NHU05396013-001.pdf		1442Kb	Adobe PDF	344	View/Open
index.html		0Kb	HTML	304	View/Open

Loading...