Catálogo de publicaciones - libros

Various similarity measures have been proposed for ontology integration to identify and suggest possible matches of components in a semi-automatic process. A (basic) Multi Match Algorithm (MMA) can be used to combine these measures effectively, thus making it easier for users in such applications to identify “ideal” matches found. We propose a multi-level extension of MMA, called MLMA, which assumes the collection of similarity measures are partitioned by the user, and that there is a partial order on the partitions, also defined by the user. We have developed a running prototype of the proposed multi level method and illustrate how our method yields improved match results compared to the basic MMA. While our objective in this study has been to develop tools and techniques to support the hybrid approach we introduced earlier for ontology integration, the ideas can be applied in information sharing and ontology integration applications.

Semantic Interoperability is a major issue for National Spatial data Infrastructures (NSDIs) and mapping across heterogeneous databases is essential for such interoperability. Mapping of schemas based on ontology mapping provides opportunities for semantic translation of schemas elements and hence for database queries across heterogeneous sources. Such semantics based mappings are usually human centered processes. This paper demonstrates semi-automatic mapping using semantic similarity values from an electronic lexicon. Lexical similarity of class names and class structures constitute knowledge base for mapping between two schemas. We employ semantic mapping based on synonym similarity matches from WordNet. We use heuristics based propagation of similarities using attribute mapping and superclass-subclass relations. The machine based similarity values are seen to be comparable to human generated values of mapping.

There are many different kinds of ontologies used for different purposes in modern computing. A continuum exists from lightweight ontologies to formal ontologies. In this paper we compare and contrast the lightweight ontology and the formal ontology approaches to data interoperability. Both approaches have strengths and weaknesses, but they both lack scalability because of the problem. We present an approach that combines their strengths and avoids their weaknesses. In this approach, the ontology includes only high level concepts; subtle differences in the interpretation of the concepts are captured as context descriptions outside the ontology. The resulting ontology is simple, thus it is easy to create. It also provides a structure for context descriptions. The structure can be exploited to facilitate automatic composition of context mappings. This mechanism leads to a scalable solution to semantic interoperability among disparate data sources and contexts.

The growth and variety of distributed information sources imply a need to exchange and/or to share information extracted from various and heterogeneous databases. The cooperation of heterogeneous information systems requires advanced architectures able to solve conflicts coming from data heterogeneity (structural and semantic heterogeneity). To resolve semantic conflicts relatively to evolutive domain ontologies following databases evolution according to the dialogue between agents, taking care of scalability issues, we propose a multi-agent system. These interaction protocols allowing ontologies evolution are currently implemented by using Java and the JADE (Java Agent DEvelopment framework) platform.

With the emergence of enterprise-wide information systems, ontologies have become by definition a valuable aid for efficient database schema modeling and integration, in addition to their use in other disciplines such as the semantic web and natural language processing. This paper presents another important utilization of ontologies in database schemas: schema evolution. Specifically, our research concentrates on a new three-layered approach for schema evolution. These three layers are 1) a schema repository, 2) a domain ontology called a , and 3) a multi-representation strategy to enable powerful change management. This a priori approach for schema evolution, in contrast with existing a posteriori solutions, can be employed for any data model and for both 1) design from scratch and evolution and 2) redesign and evolution of the database. The paper focuses on the two main foundations of this approach, the requirements ontology and the multi-representation strategy which is based on a stamping mechanism.

OLAP (On-Line Analytical Processing) operations, such as roll-up or drill-down, depend on data warehouse dimension hierarchies in order to aggregate information at different levels of detail and support the decision-making process required by final users. This is why it is crucial to capture adequate hierarchies in the requirement analysis stage. However, operational data could not be enough for supplying information to construct every level of these hierarchies. In this paper, we apply knowledge given by relationships among concepts from WordNet to overcome this problem. Therefore, richer dimension hierarchies will be specified in the data warehouse, and OLAP tools will be able to show proper information to improve decision-making process. Decision makers thus will be able to achieve their information needs for analysis. Finally, we will show the benefits of our approach by providing a case study in which a poor hierarchy is enriched with new levels of aggregation.

In this paper we propose a method for efficient management of large spatial ontologies. Current spatial ontologies are usually represented using an ontology language, such as OWL and stored as OWL files. However, we have observed some shortcomings using this approach especially in the efficiency of spatial query processing. This fact motivated the development of a hybrid approach that uses an R-tree as a spatial index structure. In this way we are able to support efficient query processing over large spatial ontologies, maintaining the benefits of ontological reasoning. We present a case study for emergency teams during Search and Rescue (SaR) operations showing how an Ontology Data Service (SHARE-ODS) can benefit from a spatial index. Performance evaluation shows the superiority of our proposed technique compared to the original approach. To the best of our knowledge, this is the first attempt to address the problem of efficient management of large spatial ontology bases.

It is a well known fact that the data mining process can generate thousands of patterns from data. Various measures exist for evaluating and ranking these discovered patterns but often they don’t consider user subjective interest. We propose an ontology-based data-mining methodology called ExCIS (Extraction using a Conceptual Information System) for integrating expert prior knowledge in a data-mining process. Its originality is to build a specific Conceptual Information System related to the application domain in order to improve datasets preparation and results interpretation. This paper focus on our ontological choices and an interestingness measure IMAK which evaluates patterns considering expert knowledge.

Desktop search tools are becoming more popular; they allow full text searches using inverted indexes. Yet, the amount of locally stored data that they have to deal with is increasing rapidly. A different approach is to analyze the semantic relationships among collected data and thus preprocess the data semantically. The goal is to allow searches based on relationships among various objects rather than focusing on objects’ names. This would allow for searches far more sophisticated than those based on full text analysis. We introduce a database architecture based on an existing software prototype that is capable of meeting the various demands of a semantic information manager. This architecture is also capable of storing and querying RDF and RDF schemata. Moreover, RDF is used as a key part of the technology. Therefore, in this scenario, RDF is used not only to enrich the Web with machine-processable semantics, but also to incorporate it into a kind of Semantic Desktop Search Engine. In this paper, we describe the underlying technology of this research project.