Iranian journal of fuzzy systems
Volume:2 Issue: 2, 2006

The proposed IAFC neural networks have both the stability and the plasticity because they use the control structure which is similar to that of the ART-1(Adaptive Resonance Theory) neural network. The unsupervised IAFC neural network is the unsupervised neural network which uses the fuzzy leaky learning rule. This fuzzy leaky learning rule controls the updating amounts by membership values. The supervised IAFC neural networks are the supervised neural networks which use the fuzzified versions of Learning Vector Quantization(LVQ). In this paper, several important adaptive learning algorithms are also compared in the viewpoint of structure and learning rule. The performances of several adaptive learning algorithms are compared using Iris data.

In this paper, a pointwise pseudo-metric function on the L-real line is constructed. It is proved that the topology induced by this pointwise pseudo-metric is the usual topology.

In this paper, we deal with fuzzy random variables for inputs and outputs in Data Envelopment Analysis (DEA). These variables are considered as fuzzy random flat LR numbers with known distribution. The problem is to find a method for converting the imprecise chance-constrained DEA model into a crisp one. This can be done by first, defuzzification of imprecise probability by constructing suitable membership function, second, defuzzification of the parameters which has been done using an α-cut and finally, converting the chance-constrained DEA into a crisp model using the method of Cooper [4].

A theorem was recently introduced to establish a relationship between goal programming and fuzzy programming for vectormaximum problems. In this short note it is shown that the relationship does not exist under all circumstances. The necessary correction is proposed.

In this paper we present a method for solving fuzzy linear systems by two crisp linear systems. Also necessary and sufficient conditions for existence of solution are given. Some numerical examples illustrate the efficiency of the method

This paper presents the application of fuzzy-neuro method to investigate transformer inrush current. Recently, the frequency environment of power systems has been made more complicated and the magnitude of second harmonic in inrush current has been decreased because of the improvement of cast steel. Therefore, traditional approaches will likely mal-operate in the case of magnetizing inrush with low second component and internal faults with high second harmonic. The proposed scheme enhances the inrush detection sensitivity of conventional techniques by using a fuzzy-neuro approach. Details of the design procedure and the results of performance studies with the proposed detector are given in the paper. Performance studies’ results show that the proposed algorithm is fast and accurate.

The present article discusses and presents a new comprehensive approach aimed at measuring the maturity and quality of software processes. This method has been designed on the basis of Software Capability Maturity Model (SW-CMM) and the Multi-level Fuzzy Inference Model and is used as a measurement and analysis tool. Among the most important characteristics of this method one can mention simple usage, accuracy, quantitative measures and comparability. Fuzzy logic-based tools are designed to provide such functions.

In this paper, we apply the Biswas'' idea of anti fuzzy subgroups to ideals of near-rings. We introduce the notion of anti fuzzy ideals of near-rings, and investigate some related properties.