slugA STUDY ON MBC ALGORITHM WITH GOODNESS FUNCTION

Journal Title: International Journal of Management, IT and Engineering - Year 2011, Vol 1, Issue 3

Abstract

In Data Mining, clustering is one of the efficient techniques used to extract useful information from large quantities of data. A cluster is a collection of data objects relatively similar to one another in some respect and relatively dissimilar to the objects in other clusters. Clustering analysis is an important technique in data mining. It is a process of grouping a set of physical or abstract objects into classes of similar objects. Clustering can be viewed as unsupervised classification. Matrix Based Clustering (MBC) is a hierarchical clustering method with a goodness function based on notions of bond and inner bond that in turn involve direct and indirect link measures. MBC employs an operation close to matrix multiplication, but with a little modification. The matrix base is thought helpful for calculations on advanced parallel machinery. The goal is to achieve good clustering performance relative to other clustering methods. Real data is tested on MBC to provide useful classification information. Matrix Based Clustering algorithm (MBC) measures the “bond" of two clusters based on a goodness function which is computed via matrix manipulation that utilizes not only direct links but also indirect links between two clusters. The effectiveness of MBC is demonstrated with several data sets that contain points in 2D space, a couple of which cannot be captured by other methods such as OPTICS, CHAMELEON, or Matlab Fuzzy Clustering.

Authors and Affiliations

P. Usha Madhuri and Dr. S. P. Rajagopalan

Keywords

Related Articles

slugSurvey on Software Process Improvement and Improvement Models

Software Process Improvement is common thinking in today’s era for any organizations’ overall benefit. It starts when organization starts thinking about the change in working environment so that performance and profit...

Impact of Technological Inputs on Quality of E-learning products: A Review

As the world shifts from the industrial era to the knowledge era, elearning is gaining prominence as the preferred method for use in pedagogical education and informal training. Embracing this advance boasts numerous a...

A Study to Measure Impact of GDP & Inflation on Credit Risk Ratios of AXIS Bank

Indian banking sector has expanded in an exponential manner in the past decade offering a wide range of services to rural, urban and metropolitan areas of the country. The banking sector reform initiated by the Reserve...

An Algorithm for Land Surface Temperature Analysis of Remote Sensing Image Coverage over Dehradun District, Uttarakhand, India

Estimation of LST from remotely sensed data is nowadays usual. LST is a key parameter in the physics of land surface processes because it is involved in the energy balance as well as in the evapotranspiration and deser...

slugIMPLIMENTATION OF WIRELESS AD-HOC NETWORK USING AURDINO CONTROLLER

The main aim of the paper is to explain the working of a robot with the use of an Ad-Hoc network. This paper explains major features of the Ad –Hoc network along with the wheel rotational analysis and the DC motor spec...

Download PDF file
  • EP ID EP18041
  • DOI -
  • Views 304
  • Downloads 14

How To Cite

P. Usha Madhuri and Dr. S. P. Rajagopalan (2011). slugA STUDY ON MBC ALGORITHM WITH GOODNESS FUNCTION. International Journal of Management, IT and Engineering, 1(3), -. https://europub.co.uk./articles/-A-18041