Login| Sign Up| Help| Contact|

Patent Searching and Data


Title:
SYSTEM AND METHOD FOR ANALYZING CLUSTER RESULTS OF LARGE AMOUNTS OF DATA
Document Type and Number:
WIPO Patent Application WO/2013/151221
Kind Code:
A1
Abstract:
The present invention relates to a system and method for analyzing the cluster results of large amounts of data. The method uses an open source MapReduce framework called Hadoop in order to calculate silhouette coefficients, which are significance test indexes capable of evaluating the cluster results of large amounts of data. In order to implement same, clustered data are divided into blocks, and input splits are created for all of the blocks. Also, the created input splits are allocated to a plurality of computers, and each of the computers stores the data of the blocks included in the input splits to a memory to calculate silhouette coefficients for each record and provides the calculated silhouette coefficients to a characteristic coefficient calculator to obtain silhouette coefficients for clusters. Thus, cluster results of large amounts of data are effectively analyzed quickly and independently.

Inventors:
LEE CHAE HYUN (KR)
KIM MIN SOENG (KR)
LEE JUN SUP (KR)
Application Number:
PCT/KR2012/008986
Publication Date:
October 10, 2013
Filing Date:
October 31, 2012
Export Citation:
Click for automatic bibliography generation   Help
Assignee:
SK PLANET CO LTD (KR)
International Classes:
G06F17/00
Foreign References:
US20120054184A12012-03-01
KR20090028953A2009-03-20
US20080256230A12008-10-16
Other References:
DEAN, JEFFREY ET AL.: "MapReduce: simplified data processing on large clusters", COMMUNICATIONS OF THE ACM, January 2008 (2008-01-01), pages 107 - 113
SHIN, MI-YOUNG: "Systematic Determination of Number of Clusters Based on Input Representation Coverage", JOURNAL OF THE INSTITUTE OF ELECTRONICS ENGINEERS OF KOREA, vol. 41, no. 6, November 2004 (2004-11-01), pages 39 - 46
Attorney, Agent or Firm:
NAM & NAM WORLD PATENT & LAW (KR)
특허법인 남앤드남 (KR)
Download PDF: