|
|
Multidimensional indexing using the VA-file
Fajt, Stanislav ; Skopal, Tomáš (advisor) ; Hoksza, David (referee)
The goal of this work was to character the indexing structure VA-file and perform the huge variety of this structure examinations. In addition, I have also described a comparison between two data structures, namely VA-File and R-tree, which has been based on the numer of experiments too. In order to analyze/search inquiries by means of the data structure VA-File the algorithm VA-NOA (near optimal algorithm) has been used. In the course of testing high-level multiplicity has been preferred. I made efforts to survey a performance of the indexing structure as widely as possible and achieve a detail view of the comparison between the qualities of the VA-File and R-tree structures. The project has resulted in the implementation of the structure VA-File that has been applied to the experiments. Since this implementation has been properly described, it is possible to see the impact of each of the parameters on the indexing file generation. Furthermore, the data structure effectiveness in the course of searching is defined in terms of multidimensional range inquiries.
|
|
|
Mining XML Integrity Constraints
Fajt, Stanislav ; Nečaský, Martin (advisor) ; Svoboda, Martin (referee)
The most important integrity constraints in XML are primary keys and foreign keys. In general, keys are essential in understanding both the structure and properties of data. They provide an instrument by which va- lues from a given set of attributes uniquely identify tuples in a database. As a result, keys are important to main database operations. Since XML beco- mes lingua franca for data exchange on the web, it is widely accepted as a model of real world data. Because XML documents in general can appear in any semi-structured form, structural constraints (including keys) are often imposed on the data that are to be modified or processed These constra- ints are formally defined in a schema.Unfortunately, in spite of the obvious advantages, the presence of a schema is not mandatory and many XML do- cuments are not joined with any. Consequently, no integrity constratins are specified in those documents, neither. This thesis is mainly focused on the inference of primary and foreign keys from XML documents. 1
|
|
|
Mining XML Integrity Constraints
Fajt, Stanislav ; Nečaský, Martin (advisor) ; Svoboda, Martin (referee)
The most important integrity constraints in XML are primary keys and foreign keys. In general, keys are essential in understanding both the structure and properties of data. They provide an instrument by which va- lues from a given set of attributes uniquely identify tuples in a database. As a result, keys are important to main database operations. Since XML beco- mes lingua franca for data exchange on the web, it is widely accepted as a model of real world data. Because XML documents in general can appear in any semi-structured form, structural constraints (including keys) are often imposed on the data that are to be modified or processed These constra- ints are formally defined in a schema.Unfortunately, in spite of the obvious advantages, the presence of a schema is not mandatory and many XML do- cuments are not joined with any. Consequently, no integrity constratins are specified in those documents, neither. This thesis is mainly focused on the inference of primary and foreign keys from XML documents. 1
|
|
|
Multidimensional indexing using the VA-file
Fajt, Stanislav ; Hoksza, David (referee) ; Skopal, Tomáš (advisor)
The goal of this work was to character the indexing structure VA-file and perform the huge variety of this structure examinations. In addition, I have also described a comparison between two data structures, namely VA-File and R-tree, which has been based on the numer of experiments too. In order to analyze/search inquiries by means of the data structure VA-File the algorithm VA-NOA (near optimal algorithm) has been used. In the course of testing high-level multiplicity has been preferred. I made efforts to survey a performance of the indexing structure as widely as possible and achieve a detail view of the comparison between the qualities of the VA-File and R-tree structures. The project has resulted in the implementation of the structure VA-File that has been applied to the experiments. Since this implementation has been properly described, it is possible to see the impact of each of the parameters on the indexing file generation. Furthermore, the data structure effectiveness in the course of searching is defined in terms of multidimensional range inquiries.
|
guest ::
