Time filter

Source Type

Figielska E.,Warsaw School of Information Technology
Control and Cybernetics | Year: 2011

This paper deals with the problem of preemptive scheduling in a two-stage flowshop with parallel unrelated machines and additional renewable resources. The objective is the minimization of makespan. The problem is NP-hard. Heuristic algorithms are proposed which join the linear programming based procedures with metaheuristic algorithms: genetic, simulated annealing and tabu search algorithm. The performance of the proposed algorithms is experimentally evaluated by comparing the solutions with a lower bound on the optimal makespan. Results of a computational experiment show that these algorithms are able to produce good solutions in short computation time and that the metaheuristics significantly improve the results for the most difficult problems. Source

Krawczak M.,Polish Academy of Sciences | Krawczak M.,Warsaw School of Information Technology | Szkatula G.,Polish Academy of Sciences
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) | Year: 2012

In this paper we developed a new methodology for grouping objects described by nominal attributes. We introduced a definition of condition's domination within each pair of cluster, and next the measure of ω-distinguishability of clusters for creating a junction of clusters. The developed method is hierarchical and agglomerative one and can be characterized both by high speed of computation as well as extremely good accuracy of clustering. © 2012 Springer-Verlag Berlin Heidelberg. Source

Krawczak M.,Warsaw School of Information Technology | Krawczak M.,Polish Academy of Sciences | Szkatula G.,Polish Academy of Sciences
Information Sciences | Year: 2015

A comparison of two objects may be viewed as an attempt to determine the degree to which they are similar or different in a given sense. Defining a good measure of proximity, or else similarity or dissimilarity between objects is very important in practical tasks as well as theoretical achievements. Each object is usually represented as a point in Cartesian coordinates, and therefore the distance between points reflects similarities between respective objects. In general, the space is assumed to be Euclidean, and a distance between points assigns a nonnegative number. From another point of view the concept of symmetry underlies essentially all theoretical treatments of similarity. Tversky (1977) provided empirical evidence of asymmetric similarities and argued that similarity should not be treated as a symmetric relation. According to Tversky's consideration, an object is described by sets of features instead of geometric points in a metric space. In this paper we propose a new measure of remoteness between sets of nominal values. Instead of considering distance between two sets, we introduce the measures of perturbation type 1 of one set by another. The consideration is based on set-theoretic operations and the proposed measure describes changes of the second set after adding the first set to it, or vice versa. The measure of sets' perturbation returns a value from [0, 1], and it must be emphasized that this measure is not symmetric in general. The difference between 1 and the sum of these two measures of perturbation of a pair of sets can be understood as Jaccard's extended similarity measure. In this paper several mathematical properties of the measure of sets' perturbation are studied, and interpretation of proximity is explained by the comparison of selected measures. © 2015 Elsevier Inc. All rights reserved. Source

Krawczak M.,Warsaw School of Information Technology
Studies in Computational Intelligence | Year: 2013

In this book, the natural structure of multilayer neural networks was used in order to consider this class of neural networks from the systems point of view: the aggregated neurons lying within one layer constitute the stage of the system and the bordering stages exchange the information on their states. The outputs of the neurons, from the same layer, constitute the state vectors. Connection weights between the neurons of the same layer are arranged in vectors and are treated as controls. In this way, we developed the interpretation of the multilayer neural networks as the multistage control systems © Springer International Publishing Switzerland 2013. Source

Krawczak M.,Warsaw School of Information Technology | Szkatula G.,Polish Academy of Sciences
Information Sciences | Year: 2014

Many methods of dimensionality reduction of data series (time series) have been introduced over the past decades. Some of them rely on a symbolic representation of the original data, however in this case the obtained dimensionality reduction is not substantial. In this paper, we introduce a new approach referred to as Symbolic Essential Attributes Approximation (SEAA) to reduce the dimensionality of multidimensional time series. In such a way we form a new nominal representation of the original data series. The approach is based on the concept of data series envelopes and essential attributes generated by a multilayer neural network. The real-valued attributes are discretized, and in this way symbolic data series representation is formed. The SEAA generates a vector of nominal values of new attributes which form the compressed representation of original data series. The nominal attributes are synthetic, and while not being directly interpretable, they still retain important features of the original data series. A validation of usefulness of the proposed dimensionality reduction is carried out for classification and clustering tasks. The experiments have shown that even for a significant reduction of dimensionality, the new representation retains information about the data series sufficient for classification and clustering of the time series. © 2013 Elsevier Inc. All rights reserved. Source

Discover hidden collaborations