International Journal of Uncertainty, Fuzziness and Knowledge-Based Systems

This paper studies a new strategy for selecting portfolios in the stock market. The strategy is inspired by two streams of previous work: (1) work on universalization of strategies for portfolio selection, which began with Thomas Cover's work on constant rebalanced portfolios, published in 1991,⁴ and (2) more general work on universalization of online algorithms,^17,21,23,30 especially Vladimir Vovk's work on the aggregating algorithm and Markov switching strategies.³² The proposed investment strategy achieves asymptotically the same exponential rate of growth as the portfolio that turns out to be best expost in the long run and does not require any underlying statistical assumptions on the nature of the stock market.

Many databases in real life involve pairs of the form (item,quantity). This kind of data can be characterized using the theory of bags. We present a general framework for extracting useful knowledge from bag databases using different types of patterns. Here we present different approaches for the task of discovering fuzzy association rules and gradual dependencies in bag databases.

In this paper we present a many-valued memory circuit based on a generalization of the R-S memory circuit known from the two-valued logical circuits.

Based on this many-valued R-S memory circuit we introduce a many-valued level-controlled memory circuit and a many-valued edge-controlled memory circuit of type “master-slave”.

We discuss problems of the stability of the circuit with respect to small errors of input and output signals as well as to small errors of the gates. We show that the stability can be preserved by restricting the set of logical values, i.e. the interval [0, 1], to a finite discrete subset of this interval.

We find the set of many-valued operations which are suitable for the implementation of the many-valued R-S memory circuit.

In this paper we deal with the entropy of fuzzy sets. We first review several defined entropies of fuzzy sets and then propose a new one. Some comparisons are made with some existing entropies to show the effectiveness of the proposed one.

The Intuitionistic Fuzzy Sets (IFSs), originated by Atanassov [1], is a useful tool to deal with vagueness and ambiguity. In the short time since their first appearance, many different distance and similarity measures of IFSs have been proposed, but they are scattered through the literature. In this paper, we give a comprehensive overview of distance and similarity measures of IFSs. Based on the weighted Hamming distance, the weighted Euclidean distance, and the weighted Hausdorff distance, respectively, we define some continuous distance and similarity measures for IFSs. We also utilize geometric distance model to define some continuous distance and similarity measures for IFSs, which are the various combinations and generalizations of the weighted Hamming distance, the weighted Euclidean distance and the weighted Hausdorff distance. Then we extend these distance and similarity measures for Interval-Valued Intuitionistic Fuzzy Sets (IVIFSs).

There exists little investigation on multiattribute decision making under intuitionistic fuzzy environments although both crisp and fuzzy multiattribute decision making have achieved a great progress. In this paper, multiattribute decision making problems using intuitionistic fuzzy sets are investigated and the TOPSIS is further extended to develop one new methodology for solving such problems. In this methodology, an interval fractional programming model is constructed on the basis of the relative closeness coefficient using the TOPSIS. Comprehensive evaluation of each alternative, which may be described as an intuitionistic fuzzy set or interval number, is calculated using two auxiliary mathematical programming problems derived from the interval fractional programming model proposed in this paper. Optimal degrees of membership for alternatives are calculated to determine their ranking order using the concept of likelihood based on the ranking method of interval numbers. Implementation process of the method proposed in this paper is illustrated with a numerical example.

Some neural network related methods have been applied to nonlinear fuzzy regression analysis by several investigators. The performance of these methods will significantly worsen when the outliers exist in the training data set. In this paper, we propose a training algorithm for fuzzy neural networks with general fuzzy number weights, biases, inputs and outputs for computation of nonlinear fuzzy regression models. First, we define a cost function that is based on the concept of possibility of fuzzy equality between the fuzzy output of fuzzy neural network and the corresponding fuzzy target. Next, a training algorithm is derived from the cost function in a similar manner as the back-propagation algorithm. Last, we examine the ability of our approach by computer simulations on numerical examples. Simulation results show that the proposed algorithm is able to reduce the outlier effects.

This section is maintained by Vladik Kreinovich. Please send your abstracts (or copies of papers that you want to see reviewed here) to vladik@utep.edu, or by regular mail to: V. Kreinovich, Department of Computer Science, University of Texas, El Paso, TX 79968, USA.