Moti Schneider

Automatic construction of FCMs

In this paper we describe a method for automatically constructing fuzzy cognitive maps based on the user provided data. This method consists of finding the degree of similarity between any two variables (represented by numerical vectors), finding whether the relation between variables is direct or inverse, and with the use of the fuzzy expert system tool (FEST) it determines the causality among variables. First, we describe the algorithm to construct automatically the FCM, and then we provide a case study to illustrate the functioning of our algorithm.

Feature-based fuzzy classification for interpretation of mammograms

Abstract Methods in fuzzy logic have been applied to serve as secondary classifier for a hierarchical classification model. The use of this model in interpretation of mammograms is discussed. Also is discussed, the inevitability of using a fuzzy approach in the problem. Finally, the two different fuzzy approaches for secondary classification are compared on basis of their performance as far as clustering is concerned. The idea of using a fuzzy covariance matrix [5,6] in the distance metric of the classical c-means algorithm [1–3] has also been tried.

The use of weighted fuzzy expected value (WFEV) in fuzzy expert systems

Abstract The use of the fuzzy expected value (FEV) as a ‘typical’ grade of membership within a fuzzy set, may occasionally generate improper results. In these cases we replace it by a new quantity — the weighted fuzzy expected value (WFEV) and demonstrate its improved performance. The applicability of this concept to decision-making in fuzzy intelligent systems is discussed and illustrated.

Design issues in fuzzy case-based reasoning

This paper focuses on the aspect of case representation in case-based reasoning systems. We introduce an augmentation in which each case is weighted individually over an independent set of properties. In addition, we encourage the use of fuzzy-valued properties, and show their incorporation in this representation. Combining these two characteristics, we receive the framework for a versatile Fuzzy CBR system.

Constructing fuzzy cognitive maps

This paper describes the method for automatically constructing fuzzy cognitive maps based on the user-provided data. This method consists of finding the degree of similarity between any two variables (represented by numerical vectors), finding whether the relation between variables is direct or inverse, and with the help of the fuzzy expert system tool (FEST) it establishes the causality among variables.<<ETX>>

A stochastic model of actions and plans for anytime planning under uncertainty

Building planning systems that operate in real domains requires coping with both uncertainty and time pressure. This article describes a model of reaction plans, which are generated using a formalization of actions and of state descriptions in probabilistic logic, as a basis for anytime planning under uncertainty. the model has the following main features. At the action level, we handle incomplete and ambiguous domain information, and reason about alternative action effects whose probabilities are given. On this basis, we generate reaction plans that specify different courses of action, reflecting the domain uncertainty and alternative action effects; if generation time was insufficient, these plans may be left unfinished, but they can be reused, incrementally improved, and finished later. At the planning level, we develop a framework for measuring the quality of plans that takes domain uncertainty and probabilistic information into account using Markov chain theory; based on this framework, one can design anytime algorithms focusing on those parts of an unfinished plan first, whose completion promises the most “gain”. Finally, the plan quality can be updated during execution, according to additional information acquired, and can therefore be used for on‐line planning.

Properties of the fuzzy expected value and the fuzzy expected interval in fuzzy environment

Abstract The fuzzy expected value (FEV) over some domain measures the typical value of some object in that domain. The evaluation of FEV requires a complete knowledge about the domain of the evaluation, and the distribution of the population in that domain [1]. Since it is not always possible to assume a complete knowledge about the domain, it is necessary to find some relaxations to the restrictions involving the evaluation of FEV. In this paper some solutions to this problem are proposed via the concept of the fuzzy expected interval (FEI).

On the use of fuzzy sets in histogram equalization

Abstract The main objective of image enhancement is to generate a new image such that the new image is more suitable for an application than the original image. One of the methods for image enhancement is histogram equalization. In this paper, an algorithm for histogram equalization is presented, based on probability. Then a new algorithm based on fuzzy sets is presented. The new method, denoted by fuzzy histogram equalization (FHE), generates images which are sharper than the images produced by the classical approach of histogram equalization.

Automatic knowledge acquisition for expert systems

It is suggested that if an expert system truly lives up to its name and given some basic, observable information on a domains structure and characteristics, the system should be able to perform its inferences to a high degree of confidence and, most important, learn from its inferences to refine them as it works more and more in the domain. An approach to automatic knowledge acquisition that attempts to achieve such a goal is presented. In order to focus on the philosophy and methods needed to realize this goal, a subset of expert systems was chosen. Fault isolation expert systems for electronic and/or electromechanical systems were selected for the development of this acquisition process. Target systems to be fault isolated are typically constructed of readily separable components known as the lowest replaceable units (LRUs), which are interconnected, receive a finite set of external inputs and offer a finite set of outputs. The application in industry of a user-friendly knowledge acquisition tool for use by nonexperts is definitely needed, and its cost savings could be significant.<<ETX>>

Fuzzy timed petri net definitions, properties, and applications

We present a new fuzzy timed Petri net model. Each transition firing in our model is associated with a fuzzy number; during transition, firings tokens are removed from input and added to output places. We consider the marks changing rate in each place as constant, and our performance analysis is based on the reachability state graph. Our model has fuzzy differential equations incorporated with it. Finally, we present the capabilities of the model by examples.