Academic Journals Database
Disseminating quality controlled scientific knowledge

An Advance towards Standard Utilities for Document Clustering

ADD TO MY LIST
 
Author(s): Aggadi Gnanesh | M.Sudhir Kumar | M.Sudhir Kumar

Journal: International Journal of Computer & Electronics Research
ISSN 2320-9348

Volume: 2;
Issue: 4;
Start page: 547;
Date: 2013;
Original page

Keywords: Clustering | Pattern Organization | Single-View Clustering | Multi-View Clustering | Similarity Measure

ABSTRACT
The term clustering is used in numerous research communities to explain methods for combination of unlabeled data. Clustering is useful in a number of exploratory pattern-analysis, assemblage, decision-making, and machine-learning circumstances, together with data mining, document recovery, image segmentation, and pattern organization. The main purpose of clustering is to organize objects of data into various separate clusters basically as the intra cluster in which resemblance is maximum and other type in which the inter cluster difference among them is maximum. There have been numerous clustering algorithms available every year and the efficiency of algorithms depends on the aptness of the similarity measure to the data at hand. The objects which are to be determined should be in the similar cluster at the same time the location of the points from someplace to begin this dimension should be outer surface of the cluster and this application is known as Multi viewpoint-based Similarity. Multiple viewpoints clustering provide an ability to find out important structures within the rule base by providing a method to structure both hierarchically and orthogonally. In recent times, multi-view clustering methods have been proposed to expand over conventional single-view clustering. It is possible to make use of more than one point of indication for creating new concept of identity.  
Why do you need a reservation system?      Save time & money - Smart Internet Solutions