###
*Maciej* SZAFARCZYK

"... direct tool wear measurement. The main role of tool probes in NC machine tools. The probes used now in industry and their drawbacks. The original concept of a tool probe using full strain gauge bridges for orientation of tool edges in four directions: +X,-X, +Z,-Z and for direct tool wear measuremen ..."

Abstract
- Add to MetaCart

direct tool wear measurement. The main role of tool probes in NC machine tools. The probes used now in industry and their drawbacks. The original concept of a tool probe using full strain gauge bridges for orientation of tool edges in four directions: +X,-X, +Z,-Z and for direct tool wear measurement. 1.

###
*Maciej* Piasecki, Wrocaw

, 1999

"... this paper is to construct semantic representation structures (SRS hereafter) for Polish noun phrases (NPs), covering the following issues: the proper representation of the reference and quantification aspects of NP meaning and their relationships, representation of semantic -pragmatic interface con ..."

Abstract
- Add to MetaCart

this paper is to construct semantic representation structures (SRS hereafter) for Polish noun phrases (NPs), covering the following issues: the proper representation of the reference and quantification aspects of NP meaning and their relationships, representation of semantic -pragmatic interface connected with reference and existential presupposition, and the phenomenon, characteristic of Polish (and many other Slavic languages, too) of the influence of the order of subsequent NPs in a sentence on their meanings. The problem of the possible existence of determiner category will also be discussed. The proposed SRS are intended to form a base for computer applications, and therefore, should be interpretable in a formal model

###
Grzegorz GODLEWSKI * *Maciej*

"... corpus, medical text, annotation morpho-syntactic tagging, natural language processing There is only one large corpus of Polish annotated with morpho-syntactic information, namely The IPI PAN Corpus (IPIC). This situation is a big obstacle in creation of tools for natural language processing dedicat ..."

Abstract
- Add to MetaCart

corpus, medical text, annotation morpho-syntactic tagging, natural language processing There is only one large corpus of Polish annotated with morpho-syntactic information, namely The IPI PAN Corpus (IPIC). This situation is a big obstacle in creation of tools for natural language processing dedicated to the domain of medical texts. However, the real life medical texts exhibit features making them very distinct from the most of the texts stored in IPIC. In the paper, the attempts to create a larger corpus of medical texts called KorMedIIS are presented. The subsequent phases of texts gathering, their lexical analysis and morphosyntactic annotation are discussed. The works on preparation of extended version of tools for processing natural language, namely a morphological analyser and a tagger, are presented. Different linguistic problems encountered are discussed. A special tool for the edition of the corpus, searching and correcting it is also presented. 1.

###
Carlangelo Liverani *Maciej* P. Wojtkowski

- Pacific Journal of Math
, 1994

"... . We introduce a generalization of the Hilbert projective metric to the space of positive definite matrices which we view as part of the Lagrangian Grassmannian. x0. Introduction. In his treatment of Kalman Bucy filters Bougerol [B1], [B2] uses the Riemannian metric on the set of positive definite ..."

Abstract
- Add to MetaCart

. We introduce a generalization of the Hilbert projective metric to the space of positive definite matrices which we view as part of the Lagrangian Grassmannian. x0. Introduction. In his treatment of Kalman Bucy filters Bougerol [

*B*1], [*B*2] uses the Riemannian metric on the set of positive###
Referee dr inż. *MACIEJ* ZAKRZEWICZ

, 2003

"... In this thesis we propose a description-oriented algorithm for clustering of results obtained from Web search engines called LINGO. The key idea of our method is to first discover meaningful cluster labels and then, based on the labels, determine the actual content of the groups. We show how the clu ..."

Abstract
- Add to MetaCart

In this thesis we propose a description-oriented algorithm for clustering of results obtained from Web search engines called LINGO. The key idea of our method is to first discover meaningful cluster labels and then, based on the labels, determine the actual content of the groups. We show how the cluster label discovery can be accomplished with the use of the Latent Semantic Indexing technique. We also discuss several factors that influence the quality of cluster description, such as input data preprocessing and language identification. As part of this thesis we prepared an implementation of our algorithm within the framework of Carrot 2. To evaluate the practical performance of our algorithm we propose an expert-based scheme for assessment of clustering results. Streszczenie W niniejszej pracy magisterskiej proponujemy algorytm grupowania wyników zapytań skierowanych do internetowych serwisów wyszukiwawczych, którego najważniejszym celem jest zapewnienie wysokiej jakości opisu utworzonych skupień. Podstawową ideą proponowanego algorytmu jest rozpoczęcie procesu grupowania od ustalenia zrozumiałych dla użytkownika opisów tworzonych grup. W kolejnej fazie algorytmu znalezione uprzednio opisy stają się podstawą do przydziału wyników zapytania do odpowiednich skupień. W pracy tej pokazujemy w jaki sposób technika Latent Semantic Indexing może być wykorzystania podczas poszukiwania odpowiednich opisów grup. Wskazujemy również inne elementy wpływające na jakość opisu sukupień takie jak wstępne przetwarzanie danych oraz rozpoznawanie języka w którym zostały napisane grupowane dokumenty. Częścią prezentowanej pracy jest implementacja proponowanego algorytmu w środowisku Carrot 2. Dla potrzeb oceny praktycznej wydajności stworzonego algorytmu proponujemy ekspercką metodę oceny jakości wyników grupowania.

###
*Maciej* GÓRKA, Mariusz-Orion JÊDRYSEK,

"... Min er al og i cal and ox y gen iso to pe com po si tion of in or ganic dust-fall in Wrocaw ..."

Abstract
- Add to MetaCart

Min er al og i cal and ox y gen iso to pe com po si tion of in or ganic dust-fall in Wrocaw

###
RATIONAL POINTS ON CERTAIN QUINTIC HYPERSURFACES *MACIEJ* ULAS

, 810

"... Abstract. Let f(x) = x 5 +ax 3 +bx 2 +cx ∈ Z[x] and consider the hypersurface of degree five given by the equation Vf: f(p) + f(q) = f(r) + f(s). Under the assumption b ̸ = 0 we show that there exists Q-unirational elliptic surface contained in Vf. If b = 0, a < 0 and −a ̸ ≡ 2, 18, 34 (mod 48) ..."

Abstract
- Add to MetaCart

Abstract. Let f(x) = x 5 +ax

*3*+bx 2 +cx ∈ Z[x] and consider the hypersurface of degree five given by the equation Vf: f(p) + f(q) = f(r) + f(s). Under the assumption*b*̸ = 0 we show that there exists Q-unirational elliptic surface contained in Vf. If*b*= 0, a < 0 and −a ̸ ≡ 2, 18, 34 (mod 48###
FRACTAL WEYL LAWS FOR QUANTUM RESONANCES *MACIEJ* ZWORSKI

"... We present results of recent work with Johannes Sjöstrand [18] on upper bounds of the number of semiclassical resonances for systems with chaotic classical dynamics. These upper bounds are interpreted as “fractal Weyl laws for resonances ” since the exponent is now related to the dimension of the tr ..."

Abstract
- Add to MetaCart

We present results of recent work with Johannes Sjöstrand [18] on upper bounds of the number of semiclassical resonances for systems with chaotic classical dynamics. These upper bounds are interpreted as “fractal Weyl laws for resonances ” since the exponent is now related to the dimension of the trapped set of the classical system. Despite some

###
*Maciej* Paszyński∗ h-RELATION PERSONALIZED COMMUNICATION STRATEGY

"... This paper considers the communication patterns arising from the partition of geometrical domain into sub-domains, when data is exchanged between processors assigned to adjacent sub-domains. It presents the algorithm constructing bipartite graphs covering the graph representation of the partitioned ..."

Abstract
- Add to MetaCart

This paper considers the communication patterns arising from the partition of geometrical domain into sub-domains, when data is exchanged between processors assigned to adjacent sub-domains. It presents the algorithm constructing bipartite graphs covering the graph representation of the partitioned domain, as well as the scheduling algorithm utilizing the coloring of the bipartite graphs. Specifically, when the communication pattern arises from the partition of a 2D geometric area, the planar graph representation of the domain is partitioned into not more than two bipartite graphs and a third graph with maximum vertex valency 2, by means of the presented algorithm. In the general case, the algorithm finds h − 1 or fewer bipartite graphs, where h is the maximum number of neighbors. Finally, the task of message scheduling is reduced to a set of independent scheduling problems over the bipartite graphs. The algorithms are supported by a theoretical discussion on their correctness and efficiency.

###
*Maciej* Paszyński∗ h-RELATION PERSONALIZED COMMUNICATION STRATEGY

"... This paper considers the communication patterns arising from the partition of geometrical domain into sub-domains, when data is exchanged between processors assigned to adjacent sub-domains. It presents the algorithm constructing bipartite graphs covering the graph representation of the partitioned ..."

Abstract
- Add to MetaCart

This paper considers the communication patterns arising from the partition of geometrical domain into sub-domains, when data is exchanged between processors assigned to adjacent sub-domains. It presents the algorithm constructing bipartite graphs covering the graph representation of the partitioned domain, as well as the scheduling algorithm utilizing the coloring of the bipartite graphs. Specifically, when the communication pattern arises from the partition of a 2D geometric area, the planar graph representation of the domain is partitioned into not more than two bipartite graphs and a third graph with maximum vertex valency 2, by means of the presented algorithm. In the general case, the algorithm finds h − 1 or fewer bipartite graphs, where h is the maximum number of neighbors. Finally, the task of message scheduling is reduced to a set of independent scheduling problems over the bipartite graphs. The algorithms are supported by a theoretical discussion on their correctness and efficiency.