• Refine Query
  • Source
  • Publication year
  • to
  • Language
  • 4
  • Tagged with
  • 4
  • 4
  • 4
  • 4
  • 4
  • 3
  • 2
  • 2
  • 2
  • 2
  • 2
  • 2
  • 2
  • 2
  • 2
  • About
  • The Global ETD Search service is a free service for researchers to find electronic theses and dissertations. This service is provided by the Networked Digital Library of Theses and Dissertations.
    Our metadata is collected from universities around the world. If you manage a university/consortium/country archive and want to be added, details can be found on the NDLTD website.
1

Dynamic cubing for hierarchical multidimensional data space

Ahmed, Usman 18 February 2013 (has links) (PDF)
Data warehouses are being used in many applications since quite a long time. Traditionally, new data in these warehouses is loaded through offline bulk updates which implies that latest data is not always available for analysis. This, however, is not acceptable in many modern applications (such as intelligent building, smart grid etc.) that require the latest data for decision making. These modern applications necessitate real-time fast atomic integration of incoming facts in data warehouse. Moreover, the data defining the analysis dimensions, stored in dimension tables of these warehouses, also needs to be updated in real-time, in case of any change. In this thesis, such real-time data warehouses are defined as dynamic data warehouses. We propose a data model for these dynamic data warehouses and present the concept of Hierarchical Hybrid Multidimensional Data Space (HHMDS) which constitutes of both ordered and non-ordered hierarchical dimensions. The axes of the data space are non-ordered which help their dynamic evolution without any need of reordering. We define a data grouping structure, called Minimum Bounding Space (MBS), that helps efficient data partitioning of data in the space. Various operators, relations and metrics are defined which are used for the optimization of these data partitions and the analogies among classical OLAP concepts and the HHMDS are defined. We propose efficient algorithms to store summarized or detailed data, in form of MBS, in a tree structure called DyTree. Algorithms for OLAP queries over the DyTree are also detailed. The nodes of DyTree, holding MBS with associated aggregated measure values, represent materialized sections of cuboids and tree as a whole is a partially materialized and indexed data cube which is maintained using online atomic incremental updates. We propose a methodology to experimentally evaluate partial data cubing techniques and a prototype implementing this methodology is developed. The prototype lets us experimentally evaluate and simulate the structure and performance of the DyTree against other solutions. An extensive study is conducted using this prototype which shows that the DyTree is an efficient and effective partial data cubing solution for a dynamic data warehousing environment.
2

Generalizing association rules in n-ary relations : application to dynamic graph analysis

Nguyen, Thi Kim Ngan 23 October 2012 (has links) (PDF)
Pattern discovery in large binary relations has been extensively studied. An emblematic success in this area concerns frequent itemset mining and its post-processing that derives association rules. In this case, we mine binary relations that encode whether some properties are satisfied or not by some objects. It is however clear that many datasets correspond to n-ary relations where n > 2. For example, adding spatial and/or temporal dimensions (location and/or time when the properties are satisfied by the objects) leads to the 4-ary relation Objects x Properties x Places x Times. Therefore, we study the generalization of association rule mining within arbitrary n-ary relations: the datasets are now Boolean tensors and not only Boolean matrices. Unlike standard rules that involve subsets of only one domain of the relation, in our setting, the head and the body of a rule can include arbitrary subsets of some selected domains. A significant contribution of this thesis concerns the design of interestingness measures for such generalized rules: besides a frequency measures, two different views on rule confidence are considered. The concept of non-redundant rules and the efficient extraction of the non-redundant rules satisfying the minimal frequency and minimal confidence constraints are also studied. To increase the subjective interestingness of rules, we then introduce disjunctions in their heads. It requires to redefine the interestingness measures again and to revisit the redundancy issues. Finally, we apply our new rule discovery techniques to dynamic relational graph analysis. Such graphs can be encoded into n-ary relations (n ≥ 3). Our use case concerns bicycle renting in the Vélo'v system (self-service bicycle renting in Lyon). It illustrates the added-value of some rules that can be computed thanks to our software prototypes.
3

Generalizing association rules in n-ary relations : application to dynamic graph analysis / Généralisation des règles d'association dans des relations n-aires : application à l'analyse de graphes dynamiques

Nguyen, Thi Kim Ngan 23 October 2012 (has links)
Le calcul de motifs dans de grandes relations binaires a été très étudié. Un succès emblématique concerne la découverte d'ensembles fréquents et leurs post-traitements pour en dériver des règles d'association. Il s'agit de calculer des motifs dans des relations binaires qui enregistrent quelles sont les propriétés satisfaites par des objets. En fait, de nombreux jeux de données se présentent naturellement comme des relations n-aires (avec n > 2). Par exemple, avec l'ajout de dimensions spatiales et/ou temporelles (lieux et/ou temps où les propriétés sont enregistrées), la relation binaire Objets x Propriétés est étendue à une relation 4-aire Objets x Propriétés x Lieux x Temps. Nous avons généralisé le concept de règle d'association dans un tel contexte multi-dimensionnel. Contrairement aux règles usuelles qui n'impliquent que des sous-ensembles d'un seul domaine de la relation, les prémisses et les conclusions de nos règles peuvent impliquer des sous-ensembles arbitraires de certains domaines. Nous avons conçu des mesures de fréquence et de confiance pour définir la sémantique de telles règles et c'est une contribution significative de cette thèse. Le calcul exhaustif de toutes les règles qui ont des fréquences et confiances suffisantes et l'élimination des règles redondantes ont été étudiés. Nous proposons ensuite d'introduire des disjonctions dans les conclusions des règles, ce qui nécessite de retravailler les définitions des mesures d'intérêt et les questions de redondance. Pour ouvrir un champ d'application original, nous considérons la découverte de règles dans des graphes relationnels dynamiques qui peuvent être codés dans des relations n-aires (n ≥ 3). Une application à l'analyse des usages de bicyclettes dans le système Vélo'v (système de Vélos en libre-service du Grand Lyon) montre quelques usages possibles des règles que nous savons calculer avec nos prototypes logiciels. / Pattern discovery in large binary relations has been extensively studied. An emblematic success in this area concerns frequent itemset mining and its post-processing that derives association rules. In this case, we mine binary relations that encode whether some properties are satisfied or not by some objects. It is however clear that many datasets correspond to n-ary relations where n > 2. For example, adding spatial and/or temporal dimensions (location and/or time when the properties are satisfied by the objects) leads to the 4-ary relation Objects x Properties x Places x Times. Therefore, we study the generalization of association rule mining within arbitrary n-ary relations: the datasets are now Boolean tensors and not only Boolean matrices. Unlike standard rules that involve subsets of only one domain of the relation, in our setting, the head and the body of a rule can include arbitrary subsets of some selected domains. A significant contribution of this thesis concerns the design of interestingness measures for such generalized rules: besides a frequency measures, two different views on rule confidence are considered. The concept of non-redundant rules and the efficient extraction of the non-redundant rules satisfying the minimal frequency and minimal confidence constraints are also studied. To increase the subjective interestingness of rules, we then introduce disjunctions in their heads. It requires to redefine the interestingness measures again and to revisit the redundancy issues. Finally, we apply our new rule discovery techniques to dynamic relational graph analysis. Such graphs can be encoded into n-ary relations (n ≥ 3). Our use case concerns bicycle renting in the Vélo'v system (self-service bicycle renting in Lyon). It illustrates the added-value of some rules that can be computed thanks to our software prototypes.
4

Dynamic cubing for hierarchical multidimensional data space / Cube de données dynamique pour un espace de données hiérarchique multidimensionnel

Ahmed, Usman 18 February 2013 (has links)
De nombreuses applications décisionnelles reposent sur des entrepôts de données. Ces entrepôts permettent le stockage de données multidimensionnelles historisées qui sont ensuite analysées grâce à des outils OLAP. Traditionnellement, les nouvelles données dans ces entrepôts sont chargées grâce à des processus d’alimentation réalisant des insertions en bloc, déclenchés périodiquement lorsque l’entrepôt est hors-ligne. Une telle stratégie implique que d’une part les données de l’entrepôt ne sont pas toujours à jour, et que d’autre part le système de décisionnel n’est pas continuellement disponible. Or cette latence n’est pas acceptable dans certaines applications modernes, tels que la surveillance de bâtiments instrumentés dits "intelligents", la gestion des risques environnementaux etc., qui exigent des données les plus récentes possible pour la prise de décision. Ces applications temps réel requièrent l’intégration rapide et atomique des nouveaux faits dans l’entrepôt de données. De plus, ce type d’applications opérant dans des environnements fortement évolutifs, les données définissant les dimensions d’analyse elles-mêmes doivent fréquemment être mises à jour. Dans cette thèse, de tels entrepôts de données sont qualifiés d’entrepôts de données dynamiques. Nous proposons un modèle de données pour ces entrepôts dynamiques et définissons un espace hiérarchique de données appelé Hierarchical Hybrid Multidimensional Data Space (HHMDS). Un HHMDS est constitué indifféremment de dimensions ordonnées et/ou non ordonnées. Les axes de l’espace de données sont non-ordonnés afin de favoriser leur évolution dynamique. Nous définissons une structure de regroupement de données, appelé Minimum Bounding Space (MBS), qui réalise le partitionnement efficace des données dans l’espace. Des opérateurs, relations et métriques sont définis pour permettre l’optimisation de ces partitions. Nous proposons des algorithmes pour stocker efficacement des données agrégées ou détaillées, sous forme de MBS, dans une structure d’arbre appelée le DyTree. Les algorithmes pour requêter le DyTree sont également fournis. Les nœuds du DyTree, contenant les MBS associés à leurs mesures agrégées, représentent des sections matérialisées de cuboïdes, et l’arbre lui-même est un hypercube partiellement matérialisé maintenu en ligne à l’aide des mises à jour incrémentielles. Nous proposons une méthodologie pour évaluer expérimentalement cette technique de matérialisation partielle ainsi qu’un prototype. Le prototype nous permet d’évaluer la structure et la performance du DyTree par rapport aux autres solutions existantes. L’étude expérimentale montre que le DyTree est une solution efficace pour la matérialisation partielle d’un cube de données dans un environnement dynamique. / Data warehouses are being used in many applications since quite a long time. Traditionally, new data in these warehouses is loaded through offline bulk updates which implies that latest data is not always available for analysis. This, however, is not acceptable in many modern applications (such as intelligent building, smart grid etc.) that require the latest data for decision making. These modern applications necessitate real-time fast atomic integration of incoming facts in data warehouse. Moreover, the data defining the analysis dimensions, stored in dimension tables of these warehouses, also needs to be updated in real-time, in case of any change. In this thesis, such real-time data warehouses are defined as dynamic data warehouses. We propose a data model for these dynamic data warehouses and present the concept of Hierarchical Hybrid Multidimensional Data Space (HHMDS) which constitutes of both ordered and non-ordered hierarchical dimensions. The axes of the data space are non-ordered which help their dynamic evolution without any need of reordering. We define a data grouping structure, called Minimum Bounding Space (MBS), that helps efficient data partitioning of data in the space. Various operators, relations and metrics are defined which are used for the optimization of these data partitions and the analogies among classical OLAP concepts and the HHMDS are defined. We propose efficient algorithms to store summarized or detailed data, in form of MBS, in a tree structure called DyTree. Algorithms for OLAP queries over the DyTree are also detailed. The nodes of DyTree, holding MBS with associated aggregated measure values, represent materialized sections of cuboids and tree as a whole is a partially materialized and indexed data cube which is maintained using online atomic incremental updates. We propose a methodology to experimentally evaluate partial data cubing techniques and a prototype implementing this methodology is developed. The prototype lets us experimentally evaluate and simulate the structure and performance of the DyTree against other solutions. An extensive study is conducted using this prototype which shows that the DyTree is an efficient and effective partial data cubing solution for a dynamic data warehousing environment.

Page generated in 0.168 seconds