Abstract
Clustering deals with huge amounts of data and aims at the discovery of their structure which becomes expressed in terms of a collection of clusters - information granules capturing the underlying topology of the data. The objective of this paper is to propose an algorithm to support clustering realized in the form of bio-soft or DNA computing. This approach is of particular interest when dealing with large and heterogeneous data sets and when being faced with an unknown number of clusters. We present the details of the algorithm of proximity clustering and show how the overall computing is supported by the individual mechanisms of DNA processing. We offer a numerical example to illustrate essential aspects of the DNA-based clustering.