Full-Text Search:
Home|Journal Papers|About CNKI|User Service|FAQ|Contact Us|中文
《Journal of Taiyuan Normal University(Natural Science Edition)》 2018-01
Add to Favorite Get Latest Update

A Clustering Algorithm for Incomplete Data Based on Dynamic Imputation

PEI Weijie;PANG Tianjie;Department of Computer Science and Technology,Taiyuan Normal University;  
Incomplete data clustering is an important problem in cluster analysis.The existing incomplete data clustering algorithm only populates the missing value once,and has not fully utilized the known information of the data,resulting in poor filling effect and affecting the effectiveness of clustering.This paper proposed an incomplete data clustering algorithm based on dynamic filling mathed.Firstly,the mean filling method is used to initialize the missing data,then,the k-means algorithm is used to cluster the populating data set,and meanwhile,the missing value is refilled by the corresponding attribute value of the class center where the missing value is located.The proposed algorithm is tested on several UCI data sets,and the results show that the algorithm is effective.
【CateGory Index】: TP311.13
Download(CAJ format) Download(PDF format)
CAJViewer7.0 supports all the CNKI file formats; AdobeReader only supports the PDF format.
©2006 Tsinghua Tongfang Knowledge Network Technology Co., Ltd.(Beijing)(TTKN) All rights reserved