Heuristic Search Technique for Clustering.

Date of Submission

December 1992

Date of Award

Winter 12-12-1993

Institute Name (Publisher)

Indian Statistical Institute

Document Type

Master's Dissertation

Degree Name

Master of Technology

Subject Name

Computer Science

Department

Electronics and Communication Sciences Unit (ECSU-Kolkata)

Supervisor

Ray, Kumar Sankar (ECSU-Kolkata; ISI)

Abstract (Summary of the Work)

The statistical problem we are concerned with is the well known clustering problem viz. to partition a set of n objects into m nonempty disjoint: subsets called clusters. This clustering or grouping is done in such a manner that within clusters a certain critera of homogenenity and be tween clusters a certain criterion of heterogenenityja maintained.Though total anumeration of all possible clustering necessarily outputs a global optimum, for large n and, m it is impossible to stick to this method. So, instead of total enumeration, the dynamic programming method for getting optimum clustering is examined which shows a considerable reduction in the number of inherent calculations. Later on we develop a heuristic function to make use of the method called dynamic programming method with reducing heuristic. The heuristic is developed in such a way that to solve the problem for small m, and n, the amount of calculation taken is almost the same but for problems with larger dimensions (1.e. larger m and n) the reduction in intermediate calculation is spactecularly well.

Comments

ProQuest Collection ID: http://gateway.proquest.com/openurl?url_ver=Z39.88-2004&rft_val_fmt=info:ofi/fmt:kev:mtx:dissertation&res_dat=xri:pqm&rft_dat=xri:pqdiss:28843360

Control Number

ISI-DISS-1992-195

Creative Commons License

Creative Commons Attribution 4.0 International License
This work is licensed under a Creative Commons Attribution 4.0 International License.

DOI

http://dspace.isical.ac.in:8080/jspui/handle/10263/6359

This document is currently not available here.

Share

COinS