文章基本信息

标题：Exploring Constraints Inconsistence for Value Decomposition and Dimension Selection Using Subspace Clustering
本地全文：下载
作者：K.Prema ; K.Sangeetha ; Dr.S.Karthik 等
期刊名称：International Journal of Innovative Research in Science, Engineering and Technology
印刷版ISSN：2347-6710
电子版ISSN：2319-8753
出版年度：2014
期号：ICETS
页码：1261
出版社：S&S Publications
摘要：The datasets which are in the form ofobject-attribute-time is referred to as threedimensional(3D) data sets. As there are manytimestamps in 3D datasets, it is very difficult tocluster. So a subspace clustering method is applied tocluster 3D data sets. Existing algorithms areinadequate to solve this clustering problem. Most ofthem are not actionable (ability to suggest profitableor beneficial action), and its 3D structure complicatesclustering process. To cluster these three-dimensional(3D) data sets a new centroid based concept isintroduced in the proposed system called PCA. ThisPCA framework is introduced to provide excellentperformance on financial and stock domain datasetsthrough the unique combination of Singular ValueDecomposition, Principle Component Analysis and3D frequent item set mining.PCA framework prunesthe entire search space to identify the significantsubspaces and clusters the datasets based on optimalcentroid value. This framework acts as theparallelization technique to tackle the space and timecomplexities.
关键词：3D supspace clustering; singular value;decomposition; numerical optimization; financial data;mining