International Journal of Uncertainty, Fuzziness and Knowledge-Based Systems

This article presents a robust fuzzy procedure for estimating change-point regression models. We propose incorporating the fuzzy change-point algorithm with the M-estimation technique for robust estimations. The fuzzy c partitions concept is embedded into the change-point regression model so the fuzzy c-regressions and fuzzy c-means clustering can be employed to obtain the estimates of change-points and regression parameters. The M estimation with a robust criterion is used to make the estimators robust to the presence of outliers and heavy-tailed distributions. We create two robust algorithms named FCH and FCT by using Huber’s and Tukey’s functions as the robust criterion respectively. Extensive experiments with numerical and real examples are provided for demonstrating the effectiveness and the superiority of the proposed algorithms. The experimental results show the proposed algorithms are resistant to atypical observations and outperform the existing methods. The proposed FCH and FCT are generally comparable but FCT performs better in the presence of extremely high leverage outliers and heavy-tailed distributions. Real data applications show the practical usefulness of the proposed method.

Over the years, the literature on individual data anonymization has burgeoned in many directions. While such diversity should be praised, it does not come without some difficulties. Currently, the task of selecting the optimal analytical environment is complicated by the multitude of available choices and the fact that the performance of any method is generally dependent of the data properties. In light of these issues, the contribution of this paper is twofold. First, based on recent insights from the literature and inspired by cryptography, it proposes a new anonymization method that shows that the task of anonymization can ultimately rely only on ranks permutations. As a result, the method offers a new way to practice data anonymization by performing it ex-ante and independently of the distributional features of the data instead of being engaged, as it is currently the case in the literature, in several ex-post evaluations and iterations to reach the protection and information properties sought after. Second, the method establishes a conceptual connection across the field, as it can mimic all the currently existing tools. To make the method operational, this paper proposes also the introduction of permutation menus in data anonymization, where recently developed universal measures of disclosure risk and information loss are used ex-ante for the calibration of permutation keys. To justify the relevance of their uses, a theoretical characterization of these measures is also proposed.

Dissimilarities are a very usual way to compare two fuzzy sets and also two interval-valued fuzzy sets. In both cases, the dissimilarity between two sets is a number. In this work, we introduce a generalization of the notion of dissimilarity for interval-valued fuzzy sets such that it assumes values on the set of subintervals instead of the set of numbers. This seems to be more realistic taking into account the available information. We also investigate its relationship with the classical notions of dissimilarity between fuzzy sets and we obtain that the new class is richer than the existing one.

Regression analysis estimates the relationships among variables which has been widely used in growth curves, and cross-validation as a model selection method assesses the generalization ability of regression models. Classical methods assume that the observation values of variables are precise numbers while in many cases data are imprecisely collected. So this paper explores the Chapman-Richards growth model which is one of the widely used growth models with imprecise observations under the framework of uncertainty theory. The least squares estimates of unknown parameters in this model are given. Moreover, cross-validation with imprecise observations is proposed. Furthermore, estimates of the expected value and variance of the uncertain error using residuals are given. In addition, ways to predict the value of response variable with new observed values of predictor variables are discussed. Finally, a numerical example illustrates our approach.

Given a set of uncertain discrete variables with a joint probability distribution and a set of observations for some of them, the most probable explanation is a set or configuration of values for non-observed variables maximizing the conditional probability of these variables given the observations. This is a hard problem which can be solved by a deletion algorithm with max marginalization, having a complexity similar to the one of computing conditional probabilities. When this approach is unfeasible, an alternative is to carry out an approximate deletion algorithm, which can be used to guide the search of the most probable explanation, by using A^* or branch and bound (the approximate+search approach). The most common approximation procedure has been the mini-bucket approach. In this paper it is shown that the use of probability trees as representation of potentials with a pruning of branches with similar values can improve the performance of this procedure. This is corroborated with an experimental study in which computation times are compared using randomly generated and benchmark Bayesian networks from UAI competitions.

In this paper, we study t-norms and t-conorms on bounded lattices. We propose new methods for generating these operators, applicable on any bounded lattice M by use of the presence of a t-norm on [0_M, k] and a t-conorm on [k, 1_M] for an element k ∊ M\{0_M, 1_M}. In addition, some corresponding examples are provided for well understanding the structure of new t-norms and t-conorms.

Classical intervals have been a very useful tool to analyze uncertain and imprecise models, in spite of operative and interpretative shortcomings. The recent introduction of modal intervals helps to overcome those limitations. In this paper, we apply modal intervals to the field of probability, including properties and axioms that form a theoretical framework applied to the Markovian analysis of Bonus-Malus systems in car insurance. We assume that the number of claims is a Poisson distribution and in order to include uncertainty in the model, the claim frequency is defined as a modal interval; therefore, the transition probabilities are modal interval probabilities. Finally, the model is exemplified through application to two different types of Bonus-Malus systems, and the attainment of uncertain long-run premiums expressed as modal intervals.

Best-Worst method (BWM) is a new multi-criteria decision-making method based on pairwise comparisons, but only the comparisons concerning the best and the worst alternatives or criteria. This method shows some significant advantages in the simplicity with a less requirement of comparison data and reliability with better consistency. This paper proposes a new consistency measure method based on the distance of the vectors of reference comparisons in BWM because the difference of the preference in two vectors directly affects the reliability of results. Through the establishment of the threshold of consistency ratio, we supplement the definition of satisfactory consistency of the comparisons in BWM. With comparisons satisfying the acceptable consistency, we use linear programming models to find all possible priority weights between the preferences given by decision maker and derive interval weights. For comparisons with unacceptable consistency, another approach is presented to find the interval weights meeting the consistent requirement. At last, several examples are used to illustrate the details of process.

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