New statistical approaches for fuzzy data

Abstract

Real observations of qualities (continuous or discrete types) are not precise numbers but more or less non-precise. The best description of such data is by so-called fuzzy numbers. The fuzziness is different from measurement errors and stochastic uncertainty. In this paper we propose definitions of fuzzy mode, fuzzy median and fuzzy mean as well as investigation of their related properties and employ these techniques in the practical applications of real life. Empirical result shows that fuzzy statistics with soft computing is more realistic and reasonable for the statistical research. Finally certain comments are suggested for the further studies. We hope that this reformation will make the corresponding fuzzy techniques more appropriate and efficient to researchers whose only experience with statistics is using the traditional methods. © World Scientific Publishing Company.