Fuzzy regression in hydrology

Abstract

A general methodology for fuzzy regression is developed and illustrated by an actual hydrological case study. Fuzzy regression may be used whenever a relationship between variables is imprecise and/or data are inaccurate and/or sample sizes are insufficient. In such cases fuzzy regression may be used as a complement or an alternative to statistical regression analysis. In fuzzy regression, several "goodness of fit" criteria may be used such as the maximum average vagueness criterion and the prediction vagueness criterion. The technique is illustrated by means of a case study involving the relationship between soil electrical resistivity and hydraulic permeability. This relationship is imprecise and based on only a few data points. In the present case a curvilinear relationship is fitted using fuzzy regression with six calculated resistivities and six measured permeabilities. Prediction vagueness criteria appears to yield a more robust fuzzy regression than the maximum average vagueness criteria. Potential application areas of fuzzy regression in hydrology are discussed further. The methodology is relatively simple, and the results can be interpreted to provide a valuable hydrological decision-making aid.

Hydrological landscape classification: investigating the performance of HAND based landscape classifications in a central European meso-scale catchment

“…The fact that transition is not sudden, reflects, similar to fuzzy set theory, the modeler's and observer's "degree of belief" (cf. Bárdossy et al, 1990) that a point belongs to a certain landscape unit, as shown in the illustrative example in Fig. 3.…”

Dissolved oxygen prediction using a possibility theory based fuzzy neural network

“…Fuzzy numbers are an extension of fuzzy set theory, and express an uncertain or imprecise quantity. These types of numbers are particularly useful for dealing with uncertainties when data are limited or imprecise (Bárdossy et al, 1990;Guyonnet et al, 2003;Huang et al, 2010;Zhang and Achari, 2010) -in other words when epistemic uncertainty exists. This type of uncertainty is in contrast to aleatory uncertainty that is typically handled using probability theory.…”

Optimising Fuzzy Neural Network Architecture for Dissolved Oxygen Prediction and Risk Analysis

“…Fuzzy numbers are an extension of fuzzy set theory [35] and express an uncertain or imprecise quantity. Fuzzy numbers are useful for dealing with uncertainties when data are limited or imprecise [36][37][38][39]. In this method, the model coefficients are defined to capture a predefined amount of observed data within different α-cut interval which are used to construct discretised fuzzy numbers.…”