Aggregation functions based on penalties

This article studies a large class of averaging aggregation functions based on minimizing a distance from the vector of inputs, or equivalently, minimizing a penalty imposed for deviations of individual inputs from the aggregated value. We provide a systematization of various types of penalty based aggregation functions, and show how many special cases arise as the result. We show how new aggregation functions can be constructed either analytically or numerically and provide many examples. We establish connection with the maximum likelihood principle, and present tools for averaging experimental noisy data with distinct noise distributions.

OWA operators in linear regression and detection of outliers

We consider the use of Ordered Weighted Averaging (OWA) in linear regression. Our goal is to replace the traditional least squares, least absolute deviation, and maximum likelihood criteria with an OWA function of the residuals. We obtain several high breakdown robust regression methods as special cases (least median, least trimmed squares, trimmed likelihood methods). We also present new formulations of regression problem. OWA-based regression is particularly useful in the presence of outliers.

Some results on Lipschitz quasi-arithmetic means

We present in this paper some properties of k-Lipschitz quasi-arithmetic means. The Lipschitz aggregation operations are stable with respect to input inaccuracies, what is a very important property for applications. Moreover, we provide sufficient conditions to determine when a quasi–arithemetic mean holds the k-Lipschitz property and allow us to calculate the Lipschitz constant k.

OWA operators in regression problems

We consider an application of fuzzy logic connectives to statistical regression. We replace the standard least squares, least absolute deviation, and maximum likelihood criteria with an ordered weighted averaging (OWA) function of the residuals. Depending on the choice of the weights, we obtain the standard regression problems, high-breakdown robust methods (least median, least trimmed squares, and trimmed likelihood methods), as well as new formulations. We present various approaches to numerical solution of such regression problems. OWA-based regression is particularly useful in the presence of outliers, and we illustrate the performance of the new methods on several instances of linear regression problems with multiple outliers.

Use of aggregation functions in decision making

A key component of many decision making processes is the aggregation step, whereby a set of numbers is summarised with a single representative value. This research showed that aggregation functions can provide a mathematical formalism to deal with issues like vagueness and uncertainty, which arise naturally in various decision contexts.

Non-monotone averaging aggregation

We advance the theory of aggregation operators and introduce non-monotone aggregation methods based on minimization of a penalty for inputs disagreements. The application in mind is processing data sets which may contain noisy values. Our aim is to filter out noise while at the same time preserve signs of unusual values. We review various methods of robust estimators of location, and then introduce a new estimator based on penalty minimisation.

Aggregation for Atanassov's intuitionistic and interval valued fuzzy sets: the median operator

Atanassov's intuitionistic fuzzy sets (AIFS) and interval valued fuzzy sets (IVFS) are two generalizations of a fuzzy set, which are equivalent mathematically although different semantically. We analyze the median aggregation operator for AIFS and IVFS. Different mathematical theories have lead to different definitions of the median operator. We look at the median from various perspectives: as an instance of the intuitionistic ordered weighted averaging operator, as a Fermat point in a plane, as a minimizer of input disagreement, and as an operation on distributive lattices. We underline several connections between these approaches and summarize essential properties of the median in different representations.

Fuzzy logic for decision support in chronic care

Computerized clinical guidelines can provide significant benefits in terms of health outcomes and costs, however, their effective computer implementation presents significant problems. Vagueness and ambiguity inherent in natural language (textual) clinical guidelines makes them problematic for formulating automated alerts or advice. Fuzzy logic allows us to formalize the treatment of vagueness in a decision support architecture. In care plan on-line (CPOL), an intranet-based chronic disease care planning system for general practitioners (GPs) in use in South Australia, we formally treat fuzziness in interpretation of quantitative data, formulation of recommendations and unequal importance of clinical indicators. We use expert judgment on cases, as well as direct estimates by experts, to optimize aggregation operators and treat heterogeneous combinations of conjunction and disjunction that are present in the natural language decision rules formulated by specialist teams.

On migrative means and copulas

In this short work we extend the results of J.Fodor and I.J. Rudas [6] characterizing migrative triangular norms, to quasi-arithmetic means. We use idempotisation construction to obtain quasi-arithmetic means migrative with respect to fixed parameter a. We also obtain the necessary and sufficient condition for a migrative triangular norm to be a copula.

On generalized bonferroni means

In this paper we investigate modeling capabilities of Bonferroni means and their generalizations. We will show that weighted Bonferroni means can model the concepts of hard and soft partial conjunction, and lead to several interesting special cases, with quite an intuitive interpretation.

Using choquet integrals for evaluating citation indices in journal ranking

The use of citation-based indices to evaluate the quality of journals is becoming increasingly widespread. Recently, ISI Web of Knowledge has begun to include three new indices in their journal statistics, including a 5-year impact factor. Here, we continue our earlier research which modeled the behavior of journal assessors based on some of these indices and the Choquet integral. We interpret the obtained fuzzy measures of many new datasets toward understanding the importance of these newly published indices and how indicative they may be of a journal's quality. The problem is one of ordinal classification, and the values of the best-fitting fuzzy measures can be obtained using the FMtools software package.

Image reduction using means on discrete product lattices

We investigate the problem of averaging values on lattices, and in particular on discrete product lattices. This problem arises in image processing when several color values given in RGB, HSL, or another coding scheme, need to be combined. We show how the arithmetic mean and the median can be constructed by minimizing appropriate penalties, and we discuss which of them coincide with the Cartesian product of the standard mean and median. We apply these functions in image processing. We present three algorithms for color image reduction based on minimizing penalty functions on discrete product lattices.