|
|
Obecné gramatiky: Normální formy a jejich aplikace
Klobučníková, Dominika ; Krčmář, Radim (oponent) ; Meduna, Alexandr (vedoucí práce)
Táto práca sa zaoberá problematikou obecných gramatík, normálnych foriem a ich aplikácií. Zameriava sa na kontextové gramatiky ako ich špeciálne prípady. Na základe analýzy tejto množiny bol navrhnutý algoritmus využívajúci princípy Cocke-Younger-Kasami algoritmu za účelom rozhodnutia, či zadaný reťazec je vetou jazyka definovaného kontextovou gramatikou. Výsledná aplikácia implementujúca toto riešenie je navrhnutá pre prácu s kontextovými gramatikami v Penttonenovej normálnej forme.
|
|
|
Demonstrace rozptýlených gramatik s jediným kontextovým pravidlem
Holas, David ; Klobučníková, Dominika (oponent) ; Křivka, Zbyněk (vedoucí práce)
Práce se zabývá tvorbou aplikace, která ze zadané frontové gramatiky konstruuje gramatiku s rozptýleným kontextem a jedním kontextovým pravidlem, derivuje v~ní zadaný řetězec a celý proces prezentuje uživateli. K derivaci řetězců byla navržena heuristika, která snižuje počet aplikovatelných pravidel v každé větné formě. Uživatel může ovlivnit průběh derivace ručním výběrem pravidel. Průběh derivace je vypisován v podobě čistého textu nebo HTML dokumentu. Byly odhaleny dvě chyby v teoretickém podkladu práce, jmenovitě v převodu frontové gramatiky do 1. normální formy a v konstrukci gramatik s rozptýleným kontextem a jedním kontextovým pravidlem. V práci se podařilo nalézt částečné řešení druhého problému, které je implementováno v aplikaci.
|
|
|
One-Sided Random Context Grammars
Zemek, Petr ; Černá, Ivana (oponent) ; Doc. In.g Petr Sosík, Dr. (oponent) ; Meduna, Alexandr (vedoucí práce)
This thesis introduces the notion of a one-sided random context grammar as a context-free-based regulated grammar, in which a set of permitting symbols and a set of forbidding symbols are attached to every rule, and its set of rules is divided into the set of left random context rules and the set of right random context rules . A left random context rule can rewrite a nonterminal if each of its permitting symbols occurs to the left of the rewritten symbol in the current sentential form while each of its forbidding symbols does not occur there. A right random context rule is applied analogically except that the symbols are examined to the right of the rewritten symbol. The thesis is divided into three parts. The first part gives a motivation behind introducing one-sided random context grammars and places all the covered material into the scientific context. Then, it gives an overview of formal language theory and some of its lesser-known areas that are needed to fully grasp some of the upcoming topics. The second part forms the heart of the thesis. It formally defines one-sided random context grammars and studies them from many points of view. Generative power, relations to other types of grammars, reduction, normal forms, leftmost derivations, generalized and parsing-related versions all belong between the studied topics. The final part of this thesis closes its discussion by adding remarks regarding its coverage. More specifically, these remarks concern application perspectives, bibliography, and open problem areas.
|
|
|
Datový a funkční model IS
Fukátko, Petr ; Držka, Štěpán (oponent) ; Koch, Miloš (vedoucí práce)
Na základě analýzy současných přístupů k informačním systémům zabývajících se turistikou byl vytvořen datový a funkční model. Datový a funkční model byl koncipován tak, aby navržený informační systém byl přínosný pro široké spektrum uživatelů a aby našel reálné uplatnění a komerční využití.
|
|
|
Formal Models of Distributed Computation
Soukup, Ondřej ; Horváth,, Géza (oponent) ; Rogalewicz, Adam (oponent) ; Meduna, Alexandr (vedoucí práce)
The present thesis introduces derivation trees for several different types of grammars in generalized Kuroda normal forms; namely, general, regular-controlled, and scattered context grammars and cooperating distributed grammar systems. It defines simple tree-based properties related to non-context-free properties of all grammars in question and shows that if there exists a limiting constant k such that every sentence in the generated language L is the frontier of a derivation tree in which the number of occurrences of the defined tree-based properties is limited by k, the language L is in fact context-free. The thesis explains that this result represents a powerful tool for showing languages to be context-free. It also provides illustrations and examples which sketch how to apply this tool in practice.
|
|
|
On Parallel Processing in Formal Models: Jumping Automata and Normal Forms
Kocman, Radim ; Černá, Ivana (oponent) ; Janoušek, Jan (oponent) ; Meduna, Alexandr (vedoucí práce)
The present thesis introduces and studies new possibilities of parallel processing in formal models. More specifically, it focuses its attention on parallel versions of jumping finite automata and on normal forms of grammars with interesting parallel properties. In the first part of this thesis, we give an initial motivation for studying parallel processing in formal models. We briefly introduce jumping models and normal forms of grammars and grammar systems. Finally, we state the precise focus and goals of our research. The second part of this thesis is focused on new results on jumping finite automata. First, we introduce n-parallel jumping finite automata that enhance the original jumping finite automaton model with multiple reading heads. The rest of the chapter then studies the accepting power of the model under two different jumping modes. Second, we introduce double-jumping finite automata and explore advanced jumping modes utilizing two heads. We study the accepting power of the models and also the closure properties of the related language families. Lastly, we introduce jumping 5'->3' Watson-Crick finite automata that combine the jumping behavior with the biology-inspired Watson-Crick finite automata that process double-stranded DNA sequences. The rest of this chapter then studies the accepting power of the model under unrestricted and various restricted conditions. The third part of this thesis is focused on new results on CD grammar systems. We introduce two types of transformations that turn arbitrary general grammars into equivalent two-component general CD grammar systems of very reduced and simplified forms. Apart from the reduction and simplification, we describe several other useful properties concerning these systems and the way they work. In the last part, we mention possible application perspectives for the introduced models and normal forms, and we conclude the thesis with the final summary and the description of theoretical perspectives for the achieved results.
|
|
|
Demonstrace rozptýlených gramatik s jediným kontextovým pravidlem
Holas, David ; Klobučníková, Dominika (oponent) ; Křivka, Zbyněk (vedoucí práce)
Práce se zabývá tvorbou aplikace, která ze zadané frontové gramatiky konstruuje gramatiku s rozptýleným kontextem a jedním kontextovým pravidlem, derivuje v~ní zadaný řetězec a celý proces prezentuje uživateli. K derivaci řetězců byla navržena heuristika, která snižuje počet aplikovatelných pravidel v každé větné formě. Uživatel může ovlivnit průběh derivace ručním výběrem pravidel. Průběh derivace je vypisován v podobě čistého textu nebo HTML dokumentu. Byly odhaleny dvě chyby v teoretickém podkladu práce, jmenovitě v převodu frontové gramatiky do 1. normální formy a v konstrukci gramatik s rozptýleným kontextem a jedním kontextovým pravidlem. V práci se podařilo nalézt částečné řešení druhého problému, které je implementováno v aplikaci.
|
|
|
Formal Models of Distributed Computation
Soukup, Ondřej ; Horváth,, Géza (oponent) ; Rogalewicz, Adam (oponent) ; Meduna, Alexandr (vedoucí práce)
The present thesis introduces derivation trees for several different types of grammars in generalized Kuroda normal forms; namely, general, regular-controlled, and scattered context grammars and cooperating distributed grammar systems. It defines simple tree-based properties related to non-context-free properties of all grammars in question and shows that if there exists a limiting constant k such that every sentence in the generated language L is the frontier of a derivation tree in which the number of occurrences of the defined tree-based properties is limited by k, the language L is in fact context-free. The thesis explains that this result represents a powerful tool for showing languages to be context-free. It also provides illustrations and examples which sketch how to apply this tool in practice.
|
|
|
Obecné gramatiky: Normální formy a jejich aplikace
Klobučníková, Dominika ; Krčmář, Radim (oponent) ; Meduna, Alexandr (vedoucí práce)
Táto práca sa zaoberá problematikou obecných gramatík, normálnych foriem a ich aplikácií. Zameriava sa na kontextové gramatiky ako ich špeciálne prípady. Na základe analýzy tejto množiny bol navrhnutý algoritmus využívajúci princípy Cocke-Younger-Kasami algoritmu za účelom rozhodnutia, či zadaný reťazec je vetou jazyka definovaného kontextovou gramatikou. Výsledná aplikácia implementujúca toto riešenie je navrhnutá pre prácu s kontextovými gramatikami v Penttonenovej normálnej forme.
|
|
|
One-Sided Random Context Grammars
Zemek, Petr ; Černá, Ivana (oponent) ; Doc. In.g Petr Sosík, Dr. (oponent) ; Meduna, Alexandr (vedoucí práce)
This thesis introduces the notion of a one-sided random context grammar as a context-free-based regulated grammar, in which a set of permitting symbols and a set of forbidding symbols are attached to every rule, and its set of rules is divided into the set of left random context rules and the set of right random context rules . A left random context rule can rewrite a nonterminal if each of its permitting symbols occurs to the left of the rewritten symbol in the current sentential form while each of its forbidding symbols does not occur there. A right random context rule is applied analogically except that the symbols are examined to the right of the rewritten symbol. The thesis is divided into three parts. The first part gives a motivation behind introducing one-sided random context grammars and places all the covered material into the scientific context. Then, it gives an overview of formal language theory and some of its lesser-known areas that are needed to fully grasp some of the upcoming topics. The second part forms the heart of the thesis. It formally defines one-sided random context grammars and studies them from many points of view. Generative power, relations to other types of grammars, reduction, normal forms, leftmost derivations, generalized and parsing-related versions all belong between the studied topics. The final part of this thesis closes its discussion by adding remarks regarding its coverage. More specifically, these remarks concern application perspectives, bibliography, and open problem areas.
|
host ::
RSS.