Vânia Maria Ponte Vidal

Towards a sound view integration methodology

View integration is investigated with the help of three classes of interrelational dependencies, inclusion dependencies, exclusion dependencies and union functional dependencies. The process of view integration is divided into two steps, combination and optimization. View combination consists in defining new interrelational dependencies that capture similarities between different views. The optimization step tries to reduce redundancy and the size of the schema. Finally, general results about interrelational dependencies are presented that lead to an optimization procedure for a restricted class of schemas.

An Ontology-Based Framework for Geographic Data Integration

Ontologies have been extensively used to model domain-specific knowledge. Recent research has applied ontologies to enhance the discovery and retrieval of geographic data in Spatial Data Infrastructures (SDIs). However, in those approaches it is assumed that all the data required for answering a query can be obtained from a single data source. In this work, we propose an ontology-based framework for the integration of geographic data. In our approach, a query posed on a domain ontology is rewritten into sub-queries submitted over multiples data sources, and the query result is obtained by the proper combination of data resulting from these sub-queries. We illustrate how our framework allows the combination of data from different sources, thus overcoming some limitations of other ontology-based approaches. Our approach is illustrated by an example from the domain of aeronautical flights.

Query Processing in a Mediator Based Framework for Linked Data Integration

In this paper, the authors present a three-level mediator based framework for linked data integration. In the approach, the mediated schema is represented by a domain ontology, which provides a conceptual representation of the application. Each relevant data source is described by a source ontology, published on the Web according to the Linked Data principles. Each source ontology is rewritten as an application ontology, whose vocabulary is restricted to be a subset of the vocabulary of the domain ontology. The main contribution of the paper is an algorithm for reformulating a user query into sub-queries over the data sources. The reformulation algorithm exploits inter-ontology links to return more complete query results. The approach is illustrated by an example of a virtual store mediating access to online booksellers.

Towards automatic generation of AXML web services for dynamic data integration

Active XML (AXML, in short) provides a declarative framework for data and service integration over the Web. In this paper, we present a framework for generation of AXML Web services for materializing the dynamic content of Web pages in data intensive Web sites. In our framework, the dynamic content of a Web page is defined as a (virtual) view, and the view is specified with the help of a set of correspondence assertions, which specifies the semantic mapping between the XML view schema and the base sources schemas. This paper focuses on an algorithm that automatically generates, based on the views correspondence assertions, the Web service that materializes the views content.

R2RML by Assertion: A Semi-automatic Tool for Generating Customised R2RML Mappings

In this paper, we demonstrate the RBA (R2RML By Assertion) tool which automatically generates customized R2RML mappings based on a set of semantic mappings that model the relationship between the relational database schema and a target ontology in RDF. The semantic mappings are specified by a set of correspondence assertions, which are simple to understand.

Revising the constraints of lightweight mediated schemas

In this article, we address the problem of changing the constraints of a mediated schema to accommodate the set of constraints of a new export schema. The relevance of this problem lies in that the constraints of a mediated schema capture the common semantics of the data sources and, as such, they must be maintained and made available to the users of the mediation environment. We first argue that such problem can be solved by computing the greatest lower bound of two theories induced by sets of constraints, defined as the intersection of the theories. Then, for an expressive family of conceptual schemas, we show how to efficiently decide logical implication and how to compute the greatest lower bound of two theories induced by sets of constraints. The family of conceptual schemas we work with partly corresponds to OWL Lite and supports the equivalent of named classes, datatype and object properties, minCardinalities and maxCardinalities, InverseFunctionalProperties, subset constraints, and disjointness constraints. Such schemas are also sufficiently expressive to encode commonly used UML constructs, such as classes, attributes, binary associations without association classes, cardinality of binary associations, multiplicity of attributes, and ISA hierarchies with disjointness, but not with complete generalizations.

A Strategy to Revise the Constraints of the Mediated Schema

In this paper, we address the problem of changing the constraints of a mediated schema M to accommodate the constraints of a new export schema E 0 . We first show how to translate the constraints of E 0 to the vocabulary of M , creating a set of constraints C 0 in such a way that the schema mapping for E 0 is correct. Then, we show how to compute the new version of the constraints of M to accommodate C 0 so that all schema mappings, including that for E 0 , are correct. We solve both problems for subset and cardinality constraints and specific families of schema mappings.

A Rigorous Approach to Schema Restructuring

A problem of schema integration is that one cannot directly merge concepts that are intuitively the same but have different representations. This problem can be solved by performing schema transformations to conform the views being integrated so that merging of classes and attributes becomes possible (schema restructuring). Previously, researchers have approached schema restructuring in an informal manner, offering heuristics to guide practitioners during integration; thus a formal underpinning for schema restructuring has not yet been established. To establish a formal underpinning, one must extend current methodologies for schema integration to be able to express the equivalence of concepts that have nonidentical representations. Such expressive power is needed to be able to formally justify the correctness of the transformations that are used during schema restructuring. Our work addresses this problem by supporting more general forms of existence dependency constraints which allow the formal definition of the most common types of transformations that are required during schema restructuring. In this paper we formally define a decomposable normal form for schemas that specifies which properties a schema should have so that certain technical problem of concept merging do not occur. We also present an algorithm to transform a schema into an equivalent schema which is in decomposable normal form.

Preserving update semantics in schema integration

In this paper, we propose a methodology for schema integration where the semantics of updates is preserved during the view integration process. We propose to divide view integration into three steps: combination, restructuring, and optimization. In the view combination step, we define the combined schema that contains all original views, plus a new set of constraints that express how data in distinct views are interrelated. The restructuring step is devoted to normalizing the views so that merging becomes possible. The optimization step tries to reduce redundancy and the size of the schema. Our methodology defines a set of transformation primitives that allows schema integration to be realized in a safe (information preserving) and algorithmic way. In the proposed transformation primitives, the relationship between the original and transformed schema is formally specified by the instance and update mappings. We introduce the notion of an update semantics preserving transformation, which guarantees that the relationships between each view and the global schema, originated during the view integration process, reflect exactly the relationships between the views as defined by the combined schema. In our approach, the view definition mappings and the view update translator can be directly defined from the instance and update mappings between the different intermediate schemas generated during the view integration process.

Efficient Maintenance of XML Views Using View Correspondence Assertions

The eXtended Markup Language (XML) has quickly emerged as the universal format for publishing and exchanging data on the Web. As a result, data sources often export XML views over base data. These views may be materialized to achieve faster data access. The main difficulty with this approach is to maintain the consistency of the materialized view with respect to changes of base data. In this paper, we propose an algorithm for the incremental maintenance of XML views. Our algorithm uses the view correspondence assertions for checking the relevance, for the view, of a base update and computes the changes needed for propagating the update to the view.