Using expert’s rules as background knowledge in the ClusDM methodology

In complex domains it is usually quite difficult to introduce context information. However, sometimes that information should be taken into account to make decisions, because it provides some relevant knowledge that cannot be expressed using an attribute-value representation. This is the case of the determination of risk of contamination of soils. In this paper, we propose to use conjunctive rules to introduce additional background knowledge to a MCDM sorting method called ClusDM. ClusDM is based on the aggregation of the data with unsupervised clustering techniques. The paper presents a new algorithm to incorporate rules to guide the clustering process in a semi-supervised way. The paper also describes how it works in the case sorting a set of possible contaminated soils, and compares the results obtained by ClusDM when rules are used or not.     

A kernel-free quadratic surface support vector machine for semi-supervised learning

In this paper, we propose a kernel-free semi-supervised quadratic surface support vector machine model for binary classification. The model is formulated as a mixed-integer programming problem, which is equivalent to a non-convex optimization problem with absolute-value constraints. Using the relaxation techniques, we derive a semi-definite programming problem for semi-supervised learning. By solving this problem, the proposed model is tested on some artificial and public benchmark data sets. Preliminary computational results indicate that the proposed method outperforms some existing well-known methods for solving semi-supervised support vector machine with a Gaussian kernel in terms of classification accuracy.     

Nonnegative splines, in particular of degree five

Summary. We investigate splines from a variational point of view, which have the following properties: (a) they interpolate given data, (b) they stay nonnegative, when the data are positive, (c) for a given integer they minimize the functional for all nonnegative, interpolating . We extend known results for to larger , in particular to and we find general necessary conditions for solutions of this restricted minimization problem. These conditions imply that solutions are splines in an augmented grid. In addition, we find that the solutions are in and consist of piecewise polynomials in with respect to the augmented grid. We find that for general, odd there will be no boundary arcs which means (nontrivial) subintervals in which the spline is identically zero. We show also that the occurrence of a boundary arc in an interval between two neighboring knots prohibits the existence of any further knot in that interval. For we show that between given neighboring interpolation knots, the augmented grid has at most two additional grid points. In the case of two interpolation knots (the local problem) we develop polynomial equations for the additional grid points which can be used directly for numerical computation. For the general (global) problem we propose an algorithm which is based on a Newton iteration for the additional grid points and which uses the local spline data as an initial guess. There are extensions to other types of constraints such as two-sided restrictions, also ones which vary from interval to interval. As an illustration several numerical examples including graphs of splines manufactured by MATLAB- and FORTRAN-programs are given. Received November 16, 1995 / Revised version received February 24, 1997     

Characterization of Stability for Cone Increasing Constraint Mappings

We investigate stability (in terms of metric regularity) for the specific class of cone increasing constraint mappings. This class is of interest in problems with additional knowledge on some nondecreasing behavior of the constraints (e.g. in chance constraints, where the occurring distribution function of some probability measure is automatically nondecreasing). It is demonstrated, how this extra information may lead to sharper characterizations. In the first part, general cone increasing constraint mappings are studied by exploiting criteria for metric regularity, as recently developed by Mordukhovich. The second part focusses on genericity investigations for global metric regularity (i.e. metric regularity at all feasible points) of nondecreasing constraints in finite dimensions. Applications to chance constraints are given.     

Incremental Model-Based Clustering for Large Datasets With Small Clusters

Clustering is often useful for analyzing and summarizing information within large datasets. Model-based clustering methods have been found to be effective for determining the number of clusters, dealing with outliers, and selecting the best clustering method in datasets that are small to moderate in size. For large datasets, current model-based clustering methods tend to be limited by memory and time requirements and the increasing difficulty of maximum likelihood estimation. They may fit too many clusters in some portions of the data and/or miss clusters containing relatively few observations. We propose an incremental approach for data that can be processed as a whole in memory, which is relatively efficient computationally and has the ability to find small clusters in large datasets. The method starts by drawing a random sample of the data, selecting and fitting a clustering model to the sample, and extending the model to the full dataset by additional EM iterations. New clusters are then added incrementally, initialized with the observations that are poorly fit by the current model. We demonstrate the effectiveness of this method by applying it to simulated data, and to image data where its performance can be assessed visually.     

On Normality of Lagrange Multipliers for State Constrained Optimal Control Problems

An optimal control problem for nonlinear ODEs, subject to mixed control-state and pure state constraints is considered. Sufficient conditions are formulated, under which unique normal Lagrange multipliers exist and are given by regular functions. These conditions include pointwise linear independence of gradients of f -active constraints and controllability of the linearized state equation. Under some additional assumptions, further regularity of the multipliers is shown.     

Robust clustering around regression lines with high density regions

Robust methods are needed to fit regression lines when outliers are present. In a clustering framework, outliers can be extreme observations, high leverage points, but also data points which lie among the groups. Outliers are also of paramount importance in the analysis of international trade data, which motivate our work, because they may provide information about anomalies like fraudulent transactions. In this paper we show that robust techniques can fail when a large proportion of non-contaminated observations fall in a small region, which is a likely occurrence in many international trade data sets. In such instances, the effect of a high-density region is so strong that it can override the benefits of trimming and other robust devices. We propose to solve the problem by sampling a much smaller subset of observations which preserves the cluster structure and retains the main outliers of the original data set. This goal is achieved by defining the retention probability of each point as an inverse function of the estimated density function for the whole data set. We motivate our proposal as a thinning operation on a point pattern generated by different components. We then apply robust clustering methods to the thinned data set for the purposes of classification and outlier detection. We show the advantages of our method both in empirical applications to international trade examples and through a simulation study.     

GEVA: geometric variability-based approaches for identifying patterns in data

This paper, arising from population studies, develops clustering algorithms for identifying patterns in data. Based on the concept of geometric variability, we have developed one polythetic-divisive and three agglomerative algorithms. The effectiveness of these procedures is shown by relating them to classical clustering algorithms. They are very general since they do not impose constraints on the type of data, so they are applicable to general (economics, ecological, genetics...) studies. Our major contributions include a rigorous formulation for novel clustering algorithms, and the discovery of new relationship between geometric variability and clustering. Finally, these novel procedures give a theoretical frame with an intuitive interpretation to some classical clustering methods to be applied with any type of data, including mixed data. These approaches are illustrated with real data on Drosophila chromosomal inversions.     

A method for estimating interpolating polynomials

The question of obtaining a lower bound for some interpolating polynomials is considered. Under specific conditions it is proved that these bounds are sharp. As a corollary of the general theorem, under specific restrictions on the points of interpolation, lower bounds for Goncharov interpolation polynomials are obtained which coincide with known upper bounds.Translated from Matematicheskie Zametki, Vol. 17, No. 4, pp. 555–561, April, 1975.     

Constraint deletion strategy in the inertia-controlling quadratic programming method

The inertia-controlling strategy in active set methods consists of choosing the working set so that the reduced Hessian never has more than one non-positive eigenvalue. Usually, this strategy has been implemented by permitting to delete constraints only at stationary points. This paper concerns the general inertia-controlling quadratic programming method, in which constraints may be deleted at non-stationary points. We consider the determination of the search direction when the reduced Hessian is positive definite, positive semidefinite and singular and indefinite or negative definite. Recurrence formulas are presented to update the search direction and multiplier estimates when the working set changes.     

Derivative reproducing properties for kernel methods in learning theory

The regularity of functions from reproducing kernel Hilbert spaces (RKHSs) is studied in the setting of learning theory. We provide a reproducing property for partial derivatives up to order s when the Mercer kernel is C^2s. For such a kernel on a general domain we show that the RKHS can be embedded into the function space C^s. These observations yield a representer theorem for regularized learning algorithms involving data for function values and gradients. Examples of Hermite learning and semi-supervised learning penalized by gradients on data are considered.     

Bayesian analysis for step‐stress accelerated life testing under progressive interval censoring

A step‐stress accelerated life testing model is considered for progressive type‐I censored experiments when the tested items are not monitored continuously but inspected at prespecified time points, producing thus grouped data. The underlying lifetime distributions belong to a general scale family of distributions. The points of stress‐level change are simultaneously inspection points as well while there is the option of assigning additional inspection points in between the stress‐level change points. In a Bayesian framework, the posterior distributions of the parameters of the model are derived for characteristic choices of prior distributions, as conjugate‐like and normal priors; vague or noninformative. The developed approach is illustrated on a simulated example and on a real data set, both known from the literature. The results are compared to previous analyses; frequentist or Bayes.     

An intuitionistic fuzzy set based S$$^$$VM model for binary classification with mislabeled information

Traditionally, robust and fuzzy support vector machine models are used to handle the binary classification problem with noise and outliers. These models in general suffer from the negative effects of having mislabeled training points and disregard position information. In this paper, we propose a novel method to better address these issues. First, we adopt the intuitionistic fuzzy set approach to detect suspectable mislabeled training points. Then we omit their labels but use their full position information to build a semi-supervised support vector machine (\(\mathrm {S^3VM}\)) model. After that, we reformulate the corresponding model into a non-convex problem and design a branch-and-bound algorithm to solve it. A new lower bound estimator is used to improve the accuracy and efficiency for binary classification. Numerical tests are conducted to compare the performances of the proposed method with other benchmark support vector machine models. The results strongly support the superior performance of the proposed method.     

Iterative processes of Fejér type in ill-posed problems with a priori information

In the latter thirty years, the solution of ill-posed problems with a priori information formed a separate field of research in the theory of ill-posed problems. We mean the class of problems, where along with the basic equation one has some additional data on the desired solution. Namely, one states some relations and constraints which contain important information on the object under consideration. As a rule, taking into account these data in a solution algorithm, one can essentially increase its accuracy for solving ill-posed (unstable) problems. It is especially important in the solution of applied problems in the case when a solution is not unique, because this approach allows one to choose a solution that meets the reality. In this paper we survey the methods for solving such problems. We briefly describe all relevant approaches (known to us), but we pay the main attention to the method proposed by us. This technique is based on the application of iterative processes of Fejér type which admit a flexible and effective realization for a wide class of a priori constraints.     

Influence analysis of non-Gaussianity by applying projection pursuit

The Gaussian distribution is the least structured from the information-theoretic point of view. In this paper, projection pursuit is used to find non-Gaussian projections to explore the clustering structure of the data. We use kurtosis as a measure of non-Gaussianity to find the projection directions. Kurtosis is well known to be sensitive to influential points/outliers, and so the projection direction will be greatly affected by unusual points. We also develop the influence functions of projection directions to investigate abnormal observations. A data example illustrates the application of these approaches.     

A note on stability of stationary points in mathematical programs with generalized complementarity constraints

We consider parameter-dependent mathematical programs with constraints governed by the generalized non-linear complementarity problem and with additional non-equilibrial constraints. We study a local behaviour of stationarity maps that assign the respective C- or M-stationarity points of the problem to the parameter. Using generalized differential calculus rules, we provide criteria for the isolated calmness and the Aubin properties of stationarity maps considered. To this end, we derive and apply formulas of some particular objects of the third-order variational analysis.     

On the Complexity of Equalizing Inequalities

We study two approaches to replace a finite mathematical programming problem with inequality constraints by a problem that contains only equality constraints. The first approach lifts the feasible set into a high-dimensional space by the introduction of quadratic slack variables. We show that then not only the number of critical points but also the topological complexity of the feasible set grow exponentially. On the other hand, the second approach bases on an interior point technique and lifts an approximation of the feasible set into a space with only one additional dimension. Here only Karush–Kuhn–Tucker points with respect to the positive and negative objective function in the original problem give rise to critical points of the smoothed problem, so that the number of critical points as well as the topological complexity can at most double.     

Some robust objectives of FCM for data analyzing

There are many data clustering techniques available to extract meaningful information from real world data, but the obtained clustering results of the available techniques, running time for the performance of clustering techniques in clustering real world data are highly important. This work is strongly felt that fuzzy clustering technique is suitable one to find meaningful information and appropriate groups into real world datasets. In fuzzy clustering the objective function controls the groups or clusters and computation parts of clustering. Hence researchers in fuzzy clustering algorithm aim is to minimize the objective function that usually has number of computation parts, like calculation of cluster prototypes, degree of membership for objects, computation part for updating and stopping algorithms. This paper introduces some new effective fuzzy objective functions with effective fuzzy parameters that can help to minimize the running time and to obtain strong meaningful information or clusters into the real world datasets. Further this paper tries to introduce new way for predicting membership, centres by minimizing the proposed new fuzzy objective functions. And experimental results of proposed algorithms are given to illustrate the effectiveness of proposed methods.     

Inertia possibilities for completions of partial hermitian matrices

A partial Hermitian matrix is one in which some entries are specified and others are considered to be free (complex) variables. Assuming the undirected graph of the specified entries is chordal, it is shown that, with certain mild restrictions, a partial Hermitian matrix may be completed to a Hermitian matrix with any inertia allowed by the specified principal submatrices through the interlacing inequalities. This generalizes earlier work dealing with the existence of positive definite completions, and. as before, the chordality assumption is, in general, necessary. Further related observations dealing with Toeplitz completions and the minimum eigenvalues of completions are also made, and these raise additional questions.     

Inertia possibilities for completions of partial hermitian matrices *

A partial Hermitian matrix is one in which some entries are specified and others are considered to be free (complex) variables. Assuming the undirected graph of the specified entries is chordal, it is shown that, with certain mild restrictions, a partial Hermitian matrix may be completed to a Hermitian matrix with any inertia allowed by the specified principal submatrices through the interlacing inequalities. This generalizes earlier work dealing with the existence of positive definite completions, and. as before, the chordality assumption is, in general, necessary. Further related observations dealing with Toeplitz completions and the minimum eigenvalues of completions are also made, and these raise additional questions.