Klaipėdos universitetas

It is important for an organization to know what capability/maturity of the process a chosen methodology could ensure. This paper is focused on DSDM Atern process maturity by CMMI.

A well-known example of global optimization that provides solutions within fixed error limits is optimization of functions with a known Lipschitz constant. In many real-life problems this constant is unknown.

The issue of making contracts between services in Service-Oriented Architecture (SOA) arises naturally, because in real life services are subjects for negotiation and contracting. The notion of contract in SOA is rather misleading, because usually means just technical specification – input and output parameters, pre- and postconditions. However, there are also other aspects of contract. Service Level Agreements (SLAs) often are used together with technical contract. SLAs describes the level of service expected by a customer from a supplier, laying out the metrics by which that service is measured, the remedies or penalties in case the agreed-upon levels are not achieved. The paper aim to clear up the notions of contract and agreement in SOA, discusses the properties of languages that are used to specify contacts and agreements, paying special attention to their legal aspects and comparing contract features in WS-Agreement ir WSLA.

Breast cancer is the most frequent women cancer form and one of the leading mortality causes among women around the world. Patients with pathological mutation of a BRCA gene have 65% lifelong breast cancer probability. It is known that such patients have different cause of illness. In this study, we have proposed a new approach for the prediction of BRCA mutation carriers by methodically applying knowledge discovery steps and utilizing data mining methods. An alternative BRCA risk assessment model has been created utilizing decision tree classifier model. The biggest challenge was a very small size and imbalanced nature of the initial dataset, which have been collected by clinicians during 4 years of clinical trial. Iterative optimization of initial dataset, optimal algorithms selection and their parameterization have resulted in higher classifier model performance, with acceptable prediction accuracy for the clinical usage. In this study, three data mining problems have been analyzed using eleven data mining algorithms.

In this paper, we present a comparative analysis of hierarchical clustering and multidimensional scaling methods for genotype and phenotype data analysis. Fisher's exact test was applied to determinate dependencies between congenital anomalies. In order to determine the relationship between the dependences of congenital anomalies, deformations, these systems’ micro anomalies and congenital anomalies associated with orofacial clefs, the Spearman and Kendall correlation coefficients were applied. It has been detected which methods are better for genetic data visualization.

The purpose of this paper is to describe the computational algorithmic generation of the high-quality digital halftones (grey/colour patterns). At the beginning, the formal model for generation of the digital halftones, the so-called grey pattern problem (GPP) is introduced. Then, the heuristic algorithm for the solution, in particular, of the grey pattern problem is discussed. Although the algorithm employed does not guarantee the optimality of the solutions found, still superior-quality, near-optimal (and in some cases probably optimal) solutions can be achieved within reasonable computation time. Further, we provide the results of the extensive computational experiments with the newly proposed, extra-large size instance (data set) of the GPP — which is the main contribution of this work. As a confirmation of the quality of the solutions produced, we also give the visual representations of several fine-looking halftone patterns and the reader can judge about the perfection of the images obtained.

The popularity of the internet has led to a very rapid growth of IPv4 (Internet Protocol v4) users. This caused a shortage of IP addresses, so it was created a new version – IPv6 (Internet Protocol v6). Currently, there are two versions of IP for IPv4 and IPv6. Due to the large differences in addressing the protocols IPv4 and IPv6 are incompatible. It is therefore necessary to find ways to move from IPv4 to IPv6. To facilitate the transition from one version to another are developed various mechanisms and strategies. Comparative analysis is done for dual stack, 6to4 tunnel and NAT64 mechanisms in this work. It has helped to reveal the shortcomings of these mechanisms and their application in selection of realization decisions.