Chantal Reynaud

Semantic integration of XML heterogeneous data sources

With the current explosion of data, retrieving and integrating information from various sources is a critical problem. The designer has to specify a mediated schema providing a homogeneous view of the sources. We report on an initial work toward automatically generating mappings between elements in the sources and the mediated schema. Information sources we are interested in are XML documents with respect to a document type definition (DTD). We describe the Xyleme project and present our approach implemented in the SAMAG system to automatically find mappings on the basis of semantic and structural criteria. Finally, we report the first results of an experiment where SAMAG has been applied to XML documents in the cultural domain.

Semantic integration in Xyleme: a uniform tree-based approach

Xyleme is a huge warehouse integrating XML data of the Web. Xyleme considers a simple data model with data trees and tree types for describing the data sources, and a simple query language based on tree queries with boolean conditions. The main components of the data model are a mediated schema modeled by an abstract tree type, as a view of a set of tree types associated with actual data trees, called concrete tree types, and a mapping expressing the connection between the mediated schema and the concrete tree types. The first contribution of this paper is formal: we provide a declarative model-theoretic semantics for Xyleme tree queries, a way of checking tree query containment, and a characterization of tree queries as a composition of branch queries. The other contributions are algorithmic and handle the potentially huge size of the mapping relation which is a crucial issue for semantic integration and query evaluation in Xyleme. First, we propose a method for pre-evaluating queries at compile time by storing some specific meta-information about the mapping into map translation tables. These map translation tables summarize the set of all the branch queries that can be generated from the mediated schema and the set of all the mappings. Then, we propose different operators and strategies for relaxing queries which, having an empty map translation table, will have no answer if they are evaluated against the data. Finally, we present a method for semi-automatically generating the mapping relation.

Alignment-based Partitioning of Large-scale Ontologies

Ontology alignment is an important task for information integration systems that can make different resources, described by various and heterogeneous ontologies, interoperate. However very large ontologies have been built in some domains such as medicine or agronomy and the challenge now lays in scaling up alignment techniques that often perform complex tasks. In this paper, we propose two partitioning methods which have been designed to take the alignment objective into account in the partitioning process as soon as possible. These methods transform the two ontologies to be aligned into two sets of blocks of a limited size. Furthermore, the elements of the two ontologies that might be aligned are grouped in a minimal set of blocks and the comparison is then enacted upon these blocks. Results of experiments performed by the two methods on various pairs of ontologies are promising.

Knowledge representation for information integration

An information integration system provides a uniform query interface to a collection of distributed and heterogeneous information sources, giving users or other agents the illusion that they interrogate a centralized and homogeneous information system. In this paper, we focus on the use of knowledge representation techniques for building mediators for information integration. A mediator is based on the specification of a single mediated schema describing a domain of interest, and on a set of source descriptions expressing how the content of each source available to the system is related to the domain of interest. These source descriptions, also called mappings because they model the correspondence between the mediated schema and the schemas of the data sources, play a central role in the query answering process. We present two recent information integration systems, namely PICSEL and Xyleme, which are illustrative of two radically different choices concerning the expressivity of the mediated schema.

Entity discovery and annotation in tables

The Web is rich of tables (e.g., HTML tables, spreadsheets, Google Fusion Tables) that host a considerable wealth of high-quality relational data. Unlike unstructured texts, tables usually favour the automatic extraction of data because of their regular structure and properties. The data extraction is usually complemented by the annotation of the table, which determines its semantics by identifying a type for each column, the relations between columns, if any, and the entities that occur in each cell. In this paper, we focus on the problem of discovering and annotating entities in tables. More specifically, we describe an algorithm that identifies the rows of a table that contain information on entities of specific types (e.g., restaurant, museum, theatre) derived from an ontology and determines the cells in which the names of those entities occur. We implemented this algorithm while developing a faceted browser over a repository of RDF data on points of interest of cities that we extracted from Google Fusion Tables. We claim that our algorithm complements the existing approaches, which annotate entities in a table based on a pre-compiled reference catalogue that lists the types of a finite set of entities; as a result, they are unable to discover and annotate entities that do not belong to the reference catalogue. Instead, we train our algorithm to look for information on previously unseen entities on the Web so as to annotate them with the correct type.

GLAM: a generic layered adaptation model for adaptive hypermedia systems

This paper introduces GLAM, a system based on situation calculus and meta-rules, which is able to provide adaptation by means of selection of actions. It is primarily designed to provide adaptive navigation. The different levels of conception, related to different aspects of the available metadata, are split in different layers in GLAM, in order to ease the conception of the adaptation system as well as to increase the potential use of complex adaptation mechanisms. GLAM uses meta-rules to handle these layers.

Modeling Information Sources for Information Integration

A critical problem in building an information mediator is to design knowledge bases describing the contents of information sources. Concepts which capture abstractions in information sources and which are usable to describe their content must be identified. This paper addresses this knowledge acquisition problem in the context of the PICSEL project, when information sources are relational databases. The main contributions are (1) semi automated techniques for identifying relevant concepts from a databases conceptual schema, and (2) a set of tools for assisting database administrators in mapping these interesting concepts on to the domain model of the PICSEL mediator.

Knowledge acquisition techniques and second-generation expert systems

We present techniques used in ADELE, a second-generation expert system (SGES), to support the knowledge acquisition activity in the diagnostic domain. The approach has been studied inside the framework of SGES; it is based on the reunions between knowledge acquisition and explanations. When new heuristic knowledge is acquired, its justifications are looked for in domain models to support the knowledge acquisition process. ADELE is a medical diagnostic reasoning system for electromyography.

Pattern-based mapping refinement

Semantic alignment between ontologies is a crucial task for information integration. There are many ongoing efforts to develop matching systems implementing various alignment techniques but it is impossible to predict what strategy is most successful for an application domain or a given pair of ontologies. Very often the quality of the results could be improved by considering the specificities of the ontologies to be aligned. In this paper, we propose a pattern-based approach implemented in the TaxoMap Framework helping an engineer to refine mappings to take into account specific conventions used in ontologies. Experiments in the topographic field within the ANR (The French National Research Agency) project GeOnto show the usefulness of such an environment both for a domain expert and an engineer, especially when the number of mappings is very large.

Using explicit ontologies to create problem solving methods

Abstract This article describes a tool called ASTREE, based on an ontology-driven approach. ASTREE has been designed to automate the construction of an application-specific model of problem-solving behaviour in cases where skeletal models cannot be reused. In ASTREE, the construction of a problem-solving model is an identification of methods capable of achieving user-specified tasks. The identification process is based on matching elements specified in the task to be achieved and elements in a set of domain knowledge delivered by an expert. This domain knowledge is called the expertise ontology because it expresses conceptualizations of domain knowledge delivered by an expert during the knowledge elicitation phase. It is argued in this article that a well-defined expertise ontology provides strong constraints for the knowledge engineering process, and in particular, for the creation of a model of problem-solving behaviour. This article presents the syntax and the semantics of the language of the expertise ontology, an extension of the Entity/Relationship Model. It then focuses on the main techniques that ASTREE uses in the identification process. An example of room assignment has been chosen to illustrate ASTREEs various techniques. A scenario leading to the construction of a task-method structure is presented at the end of the article.