Jean-François Baget

Graal: A Toolkit for Query Answering with Existential Rules

This paper presents Graal, a java toolkit dedicated to ontological query answering in the framework of existential rules. We consider knowledge bases composed of data and an ontology expressed by existential rules. The main features of Graal are the following: a basic layer that provides generic interfaces to store and query various kinds of data, forward chaining and query rewriting algorithms, structural analysis of decidability properties of a rule set, a textual format and its parser, and import of OWL 2 files. We describe in more detail the query rewriting algorithms, which rely on original techniques, and report some experiments.

Translations between RDF(S) and conceptual graphs

Though similarities between the Semantic Web language RDF(S) and languages of the Conceptual Graphs family have often been pointed out, the differences between these formalisms have been a source of difficulties while trying to translate objets of a language into the other. In this paper, we present two such transformations, that have been implemented into the CoGUI platform, and discuss their respective strengths and weaknesses.

A Pure Graph-Based Solution to the SCG-1 Initiative

This paper answers the SCG-1 initiative. The room allocation problem provided has been solved in a generic and automatic way. The solution is based on a totally declarative formal model. Basic constructs are simple graphs and the fundamental operation for doing reasonings is the graph morphism known as projection. The other formal constructs are rules and constraints defined in terms of simple graphs. The modeling framework built upon the formal model allows one to describe a problem with asserted facts, rules representing implicit knowledge about the domain, validity constraints and rules transforming the world. A prototype implementing this framework has been built upon the tool CoGITaNT. It has been used to test our modelization of the room allocation problem.

A Simulation of Co-identity with Rules in Simple and Nested Graphs

Equality of markers ajid co-reference links have always been a convenient way to denote that two concept nodes represent the same entity in conceptual graphs. This is the underlying cause of counterexamples to projection completeness with respect to these graphs FOL semantics. Several algorithms and semantics have been proposed to achieve completeness, but they do not always suit an application specific needs. In this paper, I propose to represent identity by relation nodes, which are first-class objects of the model, and I show that conceptual graphs rules can be used to represent and simulate reasonings defined by various semantics assigned to identity, be it in the case of simple or nested graphs. The interest of this method is that we can refine these rules to manage the identity needed by the application.

Default conceptual graph rules, atomic negation and Tic-Tac-Toe

In this paper, we explore the expressivity of default CG rules (a CG-oriented subset of Reiters default logics) through two applications. In the first one, we show that default CG rules provide a unifying framework for CG rules as well as polarized CGs (CGs with atomic negation). This framework allows us to study decidable subclasses of a new language mixing CG rules with atomic negation. In the second application, we use default CG rules as a formalism to model a game, an application seldom explored by the CG community. This model puts into light the conciseness provided by defaults, as well as the possibilities they offer to achieve efficient reasonings.

ALASKA for Ontology Based Data Access

Choosing the tools for the management of large and semi-structured knowledge bases has always been considered as a quite crafty task. This is due to the emergence of different solutions in a short period of time, and also to the lack of benchmarking available solutions. In this paper, we use ALASKA, a logical framework, that enables the comparison of different storage solutions at the same logical level. ALASKA translates different data representation languages such as relational databases, graph structures or RDF triples into logics. We use the platform to load semi-structured knowledge bases, store, and perform conjunctive queries over relational and non-relational storage systems.

Extending the CG Model by Simulations

Conceptual graphs (CGs) share with FOL a fundamental expressiveness limitation: only higher-order logics allow assertions of properties on predicates. This paper intends to push back this limit by reifying underlying relations of CGs (is-a , a-kind-of , referent) into first-class objects (i.e. nodes) of an equivalent, labelled graphs (LG) model.

Answering Conjunctive Regular Path Queries over Guarded Existential Rules

Ontology-mediated query answering is concerned with the problem of answering queries over knowledge bases consisting of a database instance and an ontology. While most work in the area fo-cuses on conjunctive queries (CQs), navigational queries are gaining increasing attention. In this paper, we investigate the complexity of answering two-way conjunctive regular path queries (CRPQs) over knowledge bases whose ontology is given by a set of guarded existential rules. We first consider the subclass of linear existential rules and show that CRPQ answering is EXPTIME-complete in combined complexity and NL-complete in data complexity , matching the recently established bounds for answering non-conjunctive RPQs. For guarded rules, we provide a non-trivial reduction to the linear case, which allows us to show that the complexity of CRPQ answering is the same as for CQs, namely 2EXPTIME-complete in combined complexity and PTIME-complete in data complexity.