A NEW CLUSTERING APPROACH FOR PATTERN RECOGNITION IN UNSUPERVISED ENVIRONMENT

Belur V Dasarathy

Abstract


A practical clustering technique for learning in an unsupervised mode, based on a new concept of measure of closeness of samples, has been developed. This new approach is found to be capable of zeroing on the truly inherent clusters of  the different unspecified pattern classes with a very high degree of certainty

 


Keywords


Pattern recognition; Clusters in patterns.

Full Text:

PDF

Refbacks

  • There are currently no refbacks.