DWFIST: Uma Abordagem Baseada em Data Warehouse para Exploração e Análise de Conjuntos Frequentes
Autores
4509 |
299,10,504
|
|
4510 |
299,10,504
|
|
4511 |
299,10,504
|
Informações:
Publicações do PESC
This thesis proposes the DWFIST approach, which is concerned with supporting the analysis and exploration of frequent itemsets and derived patterns, e.g. association rules, in transactional datasets. The goal of this new approach can be summarized as the following twofold contribution: provide (1) flexible pattern-retrieval capabilities without requiring the original data during the analysis phase, and (2) a standard modeling for data warehouses of frequent itemsets allowing an easier development and reuse of tools for analysis and exploration of itemset-based patterns. A data warehouse storing frequent itemsets holding on different partitions of the original transactions plays a central role in our approach. After discussing pre-processing tasks performed in the staging area, we present standard conceptual and logical schemas aiming at a standard modeling. Properties of this standard modeling allow for a flexible combination of any set of partitions. The frequent itemsets holding on any set of partitions can be retrieved along with upper and lower bounds on their frequency counts. Completeness and precision issues related to the retrieved set of frequent itemsets are discussed as well.