Granularity research based on panweighted field of pansystems

Abstract

Purpose

In order to study granularity, this paper aims to discuss how to construct granules from the view of panweighted field of pansystem.

Design/methodology/approach

By changing the panweights of panweighted field – subdivision of panweights, increase/decrease of panweights and reallocating panweights, to construct proper granules is the approach taken.

Findings

This paper provides a new method of studying granularity. If the weights of panweighted field are subdivided, then the granularity diminishes; if the panweights of panweighted field are increased, then the corresponding granularity diminished. Contrarily, the decrease of panweights of panweighted field results in the corresponding granularity increased; by reallocating panweights, use different method to construct different granules, such as compatible class, neighbor operator, compatible core, s‐s core and so on.

Research limitations/implications

How to reallocate panweights is the main limitation.

Practical implications

A very useful advice for studying granularity.

Originality/value

This paper combines the granularity with panweighted field of pansystem, and studies granularity from the view of panweighted field of pansystems.

Citation

Li, H., Lin, H., Li, Y. and Wu, A. (2008), "Granularity research based on panweighted field of pansystems", Kybernetes, Vol. 37 No. 9/10, pp. 1438-1443. https://doi.org/10.1108/03684920810907751