Catálogo de publicaciones - libros

Data warehouses and OLAP systems help to interactively analyze huge volume of data. Frequently this data contains spatial information which is useful for decision-making process. Spatial OLAP (SOLAP) refers to the integration of spatial data in multidimensional applications at physical, logical and conceptual level. Using spatial measure as a geographical object, i.e. taking in account its geometric and descriptive attributes, raises problems regarding the aggregation operation and the cube navigation in their semantic and implementation aspects. This paper defines an extended multidimensional data model which is able to support complex objects as measures, in order to handle geographical data according with its particular nature in an OLAP context. The model allows the multidimensional navigation process. OLAP operators are described which include this new concept of measure. A prototype of a SOLAP tool that handles geographical object as measures is presented.

According to increase of spatial data, many decision support systems require the fast spatio-temporal analysis. This paper proposes the improved index for efficient OLAP in a spatial data warehouse. The main idea is to use the hybrid index of the extended aggregation R-tree and the sorted hash table. The extended R-tree supports the spatial hierarchy with the level of R-tree. Also, it provides pre-aggregation for fast retrieval of the aggregated value. The sorted hash table is the transformed hash table for supporting the temporal hierarchy. So, it provides pre-aggregation of each temporal unit, year, month and etc. By the proposed hybrid index, an efficient spatio-temporal analysis can be sup-ported since it provides the spatio-temporal hierarchy and the pre-aggregated value.

In this paper we address problems of managing data warehouses (DWs) that evolve in time and we demonstrate that transactional maintenance of evolving DWs is inevitable. To this end, we propose a nested transaction model. In this model we define 5 types of transactions each of which is responsible for certain tasks. The tasks and properties of these transactions are characterized in the paper.

A widely recognized method to optimize queries invoking virtual views is query modification. The method is based on macro-substitution, where a view invocation is textually unfolded within a query and then optimized together with the query. For complex SBQL views this method is not always applicable because a view can depend on some private server resources (thus it cannot be sent to the client side) or a view is defined by a complex algorithm (thus the view cannot be unfolded within a query). For such cases we propose a reverse technique which assumes absorption of a part of the client query (a query tail) by a view definition and then, optimization of the definition. The technique has the same potential as query modification, but it is less constrained. In the paper we illustrate both techniques on examples and present some architectural peculiarities of the new method.

A hitherto unquestioned assumption made by all methods for integrity checking has been that the database satisfies its constraints before each update. This consistency assumption has been exploited for improving the efficiency of determining whether integrity is satisfied or violated after the update. Based on a notion of violation tolerance, we present and discuss an abstract property which, for any given approach to integrity checking, is an easy, sufficient condition to check whether the consistency assumption can be abandoned without sacrificing usability and efficiency of the approach. We demonstrate the usefulness of our definitions by showing that the theorem-proving approach to database integrity by Sadri and Kowalski, as well as several other well-known methods, can indeed afford to abandon the consistency assumption without losing their efficiency, while their applicability is vastly increased.

In a large modern enterprise, it is almost inevitable that different parts of an organization use different systems to produce, store, and search their critical data. In addition, the maintenance of heterogeneous database systems is becoming an overhead and a big problem for companies. Meanwhile, it is only achieved by combining the information from these various systems so that the enterprise can use the combined value of the data they contain. One way to solve this problem is to replace the legacy applications with a single information system. Introducing the new software is not enough; it is also necessary to migrate the data from the old system to the new one. As data has been collected over many years and contains a lot of information and knowledge, this most important value of a company must be preserved. Here, we present a specific solution based on a metadata repository model that we applied which addresses an organization’s operational need to integrate the custom databases of some specific hardware dependent legacy systems.

In this study, a semantic information retrieval system to access web content is proposed. Web pages existing in the web contain not only textual but also visual data. When textual and visual data are combined, the semantics of the information presented in a web page becomes richer. Consequently, types of text body and visual data are queried as one entity in a single query sentence to improve the precision, recall and r parameters of a web query. Fuzzy domain ontology to fill the gap between raw content and semantic features is used, and a model namely OAC (Object, Action and Concept) is proposed. The core of our system is the OAC Model used for fuzzy domain ontology derivation. The OAC Model serves both images and texts, equally. Several experiments are carried out on selected real web pages, and good results are obtained.

Contextual ontologies are ontologies that characterize a concept by a set of properties that vary according to context. Contextual ontologies are now crucial for users who intend to exchange information in a domain. Existing ontology languages are not capable of defining such type of ontologies. The objective of this paper is to formally define a contextual ontology language to support the development of contextual ontologies. In this paper, we use description logics as an ontology language and then we extend it by introducing a new contextual constructor.

This paper mainly focuses on proposing efficient and extensible matchmaking architecture. We ran an improved algorithm using the semantic-distance information, based on OWL-S, UDDI and semantic web architectures. Current matchmaking architectures and algorithms lack vision and they are unable to use all available information. However, our proposed architecture uses information such as path-length of the ontological tree nodes and partial results sets for composing required service even no exact match is found. Semantic-distance information may be used as selection criteria and it provides accuracy in service selection. To define concept-similarity rating by ontology managers or local users may provide a way for service selection. We can then gather second level of information other than the pre-defined match levels such as exact, subsume or plug-in. We exploit this fact in our architecture and algorithm by providing a layered and extended architecture. Different filtering layers and different specifications are applied in the matchmaking to give an extendable architecture.

The business rule approach is used in information systems to represent domain knowledge and to maintain rules systems efficiently in volatile business environment. A number of methods were proposed to develop rule models, but only few deal with reuse of knowledge acquired in the analysis of some particular domain and automatic implementation of rules. In this paper, a method for representing knowledge by ontology transformation into the rule model is described. The method is based on ontology transformation of axioms presented in a formal way into (semi-) formal information processing rules in the form of executable rules, like active DBMS triggers. The method is implemented into the developed prototype, which is described in the case study section.