Page 1 Next

Displaying 1 – 20 of 149

Showing per page

A Bimodality Test in High Dimensions

Palejev, Dean (2012)

Serdica Journal of Computing

We present a test for identifying clusters in high dimensional data based on the k-means algorithm when the null hypothesis is spherical normal. We show that projection techniques used for evaluating validity of clusters may be misleading for such data. In particular, we demonstrate that increasingly well-separated clusters are identified as the dimensionality increases, when no such clusters exist. Furthermore, in a case of true bimodality, increasing the dimensionality makes identifying the correct...

A comparative evaluation of medium- and large-scale feature selectors for pattern classifiers

Mineichi Kudo, Jack Sklansky (1998)

Kybernetika

Needs of feature selection in medium and large problems increases in many fields including medical and image processing fields. Previous comparative studies of feature selection algorithms are not satisfactory in problem size and in criterion function. In addition, no way has not shown to compare algorithms with different objectives. In this study, we propose a unified way to compare a large variety of algorithms. Our results show that the sequential floating algorithms promises for up to medium...

A learning algorithm combining functional discriminant coordinates and functional principal components

Tomasz Górecki, Mirosław Krzyśko (2014)

Discussiones Mathematicae Probability and Statistics

A new type of discriminant space for functional data is presented, combining the advantages of a functional discriminant coordinate space and a functional principal component space. In order to provide a comprehensive comparison, we conducted a set of experiments, testing effectiveness on 35 functional data sets (time series). Experiments show that constructed combined space provides a higher quality of classification of LDA method compared with component spaces.

A neuro-fuzzy system for isolated hand-written digit recognition.

Miguel Pinzolas, José Javier Astrain, Jesús Villadangos, José Ramón González de Mendívil (2001)

Mathware and Soft Computing

A neuro-fuzzy system for isolated hand-written digit recognition using a similarity fuzzy measure is presented. The system is composed of two main blocks: a first block that normalizes the input and compares it with a set of fuzzy patterns, and a second block with a multilayer perceptron to perform a neuronal classification. The comparison with the fuzzy patterns is performed via a fuzzy similarity measure that uses the Yager parametric t-norms and t-conorms. Along this work, several values of the...

A note on the computational complexity of hierarchical overlapping clustering

Mirko Křivánek (1985)

Aplikace matematiky

In this paper the computational complexity of the problem of the approximation of a given dissimilarity measure on a finite set X by a k -ultrametric on X and by a Robinson dissimilarity measure on X is investigared. It is shown that the underlying decision problems are NP-complete.

A scratch removal method

Michal Haindl, Stanislava Šimberová (1998)

Kybernetika

We present a new type of scratch removal algorithm based on a causal adaptive multidimensional prediction. The predictor use available information from the failed pixel surrounding due to spectral and spatial correlation of multispectral data but not any information from failed pixel itself. Predictor parameters cannot be directly identified so a special approximation is introduced.

A simple upper bound to the Bayes error probability for feature selection

Lorenzo Bruzzone, Sebastiano B. Serpico (1998)

Kybernetika

In this paper, feature selection in multiclass cases for classification of remote-sensing images is addressed. A criterion based on a simple upper bound to the error probability of the Bayes classifier for the minimum error is proposed. This criterion has the advantage of selecting features having a link with the error probability with a low computational load. Experiments have been carried out in order to compare the performances provided by the proposed criterion with the ones of some of the widely...

A variant of gravitational classification

Tomasz Górecki, Maciej Luczak (2014)

Biometrical Letters

In this article there is proposed a new two-parametrical variant of the gravitational classification method. We use the general idea of objects' behavior in a gravity field. Classification depends on a test object's motion in a gravity field of training points. To solve this motion problem, we use a simulation method. This classifier is compared to the 1NN method, because our method tends towards it for some parameter values. Experimental results on different data sets demonstrate an improvement...

Currently displaying 1 – 20 of 149

Page 1 Next