Abstract. The paper presents our approach to SVM implementation in parallel environment. We describe how classification learning and prediction phases were pararellised. We also propose a method for limiting the number of necessary computations during classifier construction. Our method, named one-vs-near, is an extension of typical one-vs-all approach that is used for binary classifiers to work with multiclass problems. We perform experiments of scalability and quality of the implementation. The results show that the proposed solution allows to scale up SVM that gives reasonable quality results. The proposed one-vs-near method significantly improves effectiveness of the classifier construction.

The paper presents an approach to the large scale text documents classification problem in parallel environments. A two stage classifier is proposed, based on a combination of k-nearest neighbors and support vector machines classification methods. The details of the classifier and the parallelisation of classification, learning and prediction phases are described. The classifier makes use of our method named onevs-near. It is an extension of the one-vs-all approach, typically used with binary classifiers in order to solve multiclass problems. The experiments were performed on a large scale dataset, with use of many parallel threads on a supercomputer. Results of the experiments show that the proposed classifier scales well and gives reasonable quality results. Finally, it is shown that the proposed method gives better performance compared to the traditional approach.

Public orders play an important role in the market based economic system. Currently, a model of economic development of UE countries is, to the large extent , based upon the indebtedness of private enterprises and public organizations, which creates incentives for the development of public sector. Public order's market is most dynamically developing market in Poland. The recent research (from 2012) shows, that the value of this market was estimated on 132,7 billion PLN, and it is an important source of demand for domestic products, and-if properly adjusted-might be an instrument of balanced development. A current shape of public orders is based upon 25 years of evolution, and its beginning is dated on

Fluctuations in electric fields can change the position of a gate-defined quantum dot (QD) in a semiconductor heterostructure. In the presence of magnetic field gradient, these stochastic shifts of electron's wavefunction lead to fluctuations of electron's spin splitting. The resulting spin dephasing due to charge noise limits the coherence times of spin qubits in isotopically purified Si/SiGe quantum dots. We investigate the spin splitting noise caused by such a process due to microscopic motion of charges at the semiconductor-oxide interface. We compare effects of isotropic and planar displacement of the charges and estimate their densities and typical displacement magnitudes that can reproduce experimentally observed spin splitting noise spectra. We predict that for a defect density of 1010 cm−2, visible correlations between noises in spin splitting and in energy of electron's ground state in the quantum dot are expected.