5. Clustering with Partial Supervision

  1. Witold Pedrycz1,2

Published Online: 27 JAN 2005

DOI: 10.1002/0471708607.ch5

Knowledge-Based Clustering: From Data to Information Granules

Knowledge-Based Clustering: From Data to Information Granules

How to Cite

Pedrycz, W. (2005) Clustering with Partial Supervision, in Knowledge-Based Clustering: From Data to Information Granules, John Wiley & Sons, Inc., Hoboken, NJ, USA. doi: 10.1002/0471708607.ch5

Author Information

  1. 1

    Department of Electrical and Computer Engineering, University of Alberta, Edmonton, Canada

  2. 2

    Systems Research Institute, Polish Academy of Sciences, Warsaw, Poland

Publication History

  1. Published Online: 27 JAN 2005
  2. Published Print: 7 JAN 2005

ISBN Information

Print ISBN: 9780471469667

Online ISBN: 9780471708605

SEARCH

Keywords:

  • partial supervision;
  • labeled patterns;
  • extended objective function;
  • mixture of data;
  • tracking problem

Summary

This chapter is concerned with an enhancement of the FCM with the mechanism of supervised learning conveyed in the form of a certain (usually small) portion of labeled patterns. These patterns play a role of “anchor” points for which the membership grades are given and the algorithm attempts to adhere to these values when optimizing its underlying objective function. It is shown how partial supervision enhances the performance of cluster-oriented tracking problem.