Studies in possibilistic recognition

This paper introduces an algorithm for pattern recognition. The algorithm will classify a measured object as belonging to one of N known classes or none of the classes. The algorithm makes use of fuzzy techniques and possibility is used instead of probability. The algorithm was conceived with the idea of recognizing fast moving objects, but it is shown to be more general. Fuzzy ISODATA's use as a front end to the algorithm is shown. The algorithm is shown to accomplish the objectives of correct classification or no classification. Values that describe possibility distributions are introduced with some of their properties investigated and illustrated. An expected value for a possibility distribution is also investigated. The algorithm actually proves to be adaptable to a wide variety of imprecise recognition problems. Some test results illustrate the use of the technique embodied in the algorithm and indicate its viability.     

Knowledge representation and acquisition in expert systems for pattern recognition

A new approach to the design of fuzzy expert systems is proposed. The representation of knowledge and the formation of statements by fuzzy logic tools are discussed in detail. A model of fuzzy inference is described. Primary attention is given to automatic extraction of knowledge (fuzzy inference rules) from a set of precedents. Various performance criteria for rules are introduced, and an algorithm for their generation (the method of effective restrictions) is proposed. An extension of the type of admissible rules by introducing a fuzzy disjunction operation is described. The possibility of optimizing the rules found is explored. The benefits of the approaches proposed are illustrated by experiments.     

Minimum classification error training in example based speech and pattern recognition using sparse weight matrices

The Minimum Classification Error (MCE) criterion is a well-known criterion in pattern classification systems. The aim of MCE training is to minimize the resulting classification error when trying to classify a new data set. Usually, these classification systems use some form of statistical model to describe the data. These systems usually do not work very well when this underlying model is incorrect. Speech recognition systems traditionally use Hidden Markov Models (HMM) with Gaussian (or Gaussian mixture) probability density functions as their basic model. It is well known that these models make some assumptions that are not correct. In example based approaches, these statistical models are absent and are replaced by the pure data. The absence of statistical models has created the need for parameters to model the data space accurately. For this work, we use the MCE criterion to create a system that is able to work together with this example based approach. Moreover, we extend the locally scaled distance measure with sparse, block diagonal weight matrices resulting in a better model for the data space and avoiding the computational load caused by using full matrices. We illustrate the approach with some example experiments on databases from pattern recognition and with speech recognition.     

An imputation method for dealing with missing data in regression

We develop two methods for imputing missing values in regression situations. We examine the standard fixed-effects linear-regression model Y = X β + ?, where the regressors X are fixed and ? is the error term. This research focuses on the problem of missing X values. A particular component of market-share analysis has motivated this research where the price and other promotional instruments of each brand are allowed to have their own impact on the total sales volume in a consumer-products category. When a brand is not distributed in a particular week, only a few of the many measures occurring in that observation are missing. ‘What values should be imputed for the missing measures?’ is the central question this paper addresses. This context creates a unique problem in the missing-data literature, i.e. there is no true value for the missing measure. Using influence functions, from robust statistics we develop two loss functions, each of which is a function of the missing and existing X values. These loss functions turn out to be sums of ratios of low-order polynomials. The minimization of either loss function is an unconstrained non-linear-optimization problem. The solution to this non-linear optimization leads to imputed values that have minimal influence on the estimates of the parameters of the regression model. Estimates using the method for replacing missing values are compared with estimates obtained via some conventional methods.     

Metrics of algebraic closures in pattern recognition problems with two nonoverlapping classes

It is shown that, in the pattern recognition problem with two nonoverlapping classes, the matrices of estimates of the object closeness are described by a metric. The transition to the algebraic closure of the model of recognizing operators of finite degree corresponds to the application of a special transformation of this metric. It is proved that the minimal degree correct algorithm can be found as a polynomial of a special form. A simple criterion for testing classification implementations is obtained.     

MILP approach to pattern generation in logical analysis of data

Pattern generation methods for the Logical Analysis of Data (LAD) have been term-enumerative in nature. In this paper, we present a Mixed 0-1 Integer and Linear Programming (MILP) approach that can identify LAD patterns that are optimal with respect to various previously studied and new pattern selection preferences. Via art of formulation, the MILP-based method can generate optimal patterns that also satisfy user-specified requirements on prevalence, homogeneity and complexity. Considering that MILP problems with hundreds of 0-1 variables are easily solved nowadays, the proposed method presents an efficient way of generating useful patterns for LAD. With extensive experiments on benchmark datasets, we demonstrate the utility of the MILP-based pattern generation.     

Linear programming approach to the control of discrete-time periodic systems with uncertain inputs

The problem is considered of finding a control strategy for a linear discrete-time periodic system with state and control bounds in the presence of unknown disturbances that are only known to belong to a given compact set. This kind of problem arises in practice in resource distribution systems where the demand has typically a periodic behavior, but cannot be estimated a priori without an uncertainty margin. An infinite-horizon keeping problem is formulated, which consists in confining the state within its constraint set using the allowable control, whatever the allowed disturbances may be. To face this problem, the concepts of periodically invariant set and sequence are introduced. They are used to formulate a solution strategy that solves the keeping problem. For the case of polyhedral state, control, and disturbance constraints, a computationally feasible procedure is proposed. In particular, it is shown that periodically invariant sequences may be computed off-line, and then they may be used to synthesize on-line a control strategy. Finally, an optimization criterion for the control law is discussed.     

A note on article “A tolerance approach to the fuzzy goal programming problems with unbalanced triangular membership function”

Kim and Whang use a tolerance approach for solving fuzzy goal programming problems with unbalanced membership functions [J.S. Kim, K. Whang, A tolerance approach to the fuzzy goal programming problems with unbalanced triangular membership function, European Journal of Operational Research 107 (1998) 614–624]. In this note it is shown that some results in that article are incorrect. The necessary corrections are proposed.     

A fast approach to compute fuzzy values of matrix games with payoffs of triangular fuzzy numbers

The aim of this paper is to develop an effective method for solving matrix games with payoffs of triangular fuzzy numbers (TFNs) which are arbitrary. In this method, it always assures that players’ gain-floor and loss-ceiling have a common TFN-type fuzzy value and hereby any matrix game with payoffs of TFNs has a TFN-type fuzzy value. Based on duality theorem of linear programming (LP) and the representation theorem for fuzzy sets, the mean and the lower and upper limits of the TFN-type fuzzy value are easily computed through solving the derived LP models with data taken from 1-cut set and 0-cut set of fuzzy payoffs. Hereby the TFN-type fuzzy value of any matrix game with payoffs of TFNs can be explicitly obtained. Moreover, we can easily compute the upper and lower bounds of any Alfa-cut set of the TFN-type fuzzy value for any matrix game with payoffs of TFNs and players’ optimal mixed strategies through solving the derived LP models at any specified confidence level Alfa. The proposed method in this paper is demonstrated with a numerical example and compared with other methods to show the validity, applicability and superiority.     

A finite element approach to the position problems in open-loop variable geometry trusses

The present paper looks at some kinematic and static-equilibrium problems that arise with variable-geometry trusses (VGTs). The first part of the paper looks at the use of active controls in the correction of static deformations, the second part at the position problems. The separation between deformable- and rigid-body displacements makes it possible to consider separately the corrections in each component of the structure. VGTs are considered as open-loop linkages with redundant rigid-body degrees of freedom. Owing to this redundancy, possible solutions to the inverse problem are in general infinite, for which reason it is necessary to use some optimization criteria. To tackle the problem an optimization procedure with constraints is developed for the purpose of minimizing the displacements of the actuators. Suitable use of the constraints allows us to solve the direct position problem using the same optimization procedure.     

The axiomatic approach to three values in games with coalition structure

We present a unified framework for a broad class of values in transferable utility games with coalition structure, including the Owen coalitional value and two weighted versions with weights given by the size of the coalitions. We provide three axiomatic characterizations using the properties of Efficiency, Linearity, Independence of Null Coalitions, and Coordination, with two versions of Balanced Contributions inside a Coalition and Weighted Sharing in Unanimity Games, respectively.     

Numerical approach to chaotic pattern formation in diffusive predator–prey system with Caputo fractional operator

This paper is primarily concern with the formulation and analysis of a reliable numerical method based on the novel alternating direction implicit finite difference scheme for the solution of the fractional reaction–diffusion system. In the work, the integer first‐order derivative in time is replaced with the Caputo fractional derivative operator. As a case study, the dynamics of predator–prey model is considered. In order to provide a good guidelines on the correct choice of parameters for the numerical simulation of full fractional reaction–diffusion system, its linear stability analysis is also examined. The resulting scheme is applied to solve both self‐diffusion and cross‐diffusion problems in two‐dimensions. We observed in the experimental results a range of spatiotemporal and chaotic structures that are related to Turing pattern. It was also discovered in the simulations that cross‐diffusive case gives rise to spatial patterns faster than the diffusive case. Apart from chaotic spiral‐like structures obtained in this work, it should also be mentioned that Turing patterns such as stationary spots and stripes are obtainable, depending on the initial and parameters choices.     

Resolvent estimates in W−1,p related to strongly coupled‐linear parabolic systems with coupled nonsmooth capacities

We investigate linear parabolic systems with coupled nonsmooth capacities and mixed boundary conditions. We prove generalized resolvent estimates in W^{?1, p} spaces. The method is an appropriate modification of a technique introduced by Agmon to obtain L^p estimates for resolvents of elliptic differential operators in the case of smooth boundary conditions. Moreover, we establish an existence and uniqueness result. Copyright © 2007 John Wiley & Sons, Ltd.