Search

International audience; Decision systems usually rely on a set of business rules that describe the expected behavior of a system or an organization and that determine the decisions to be taken in different situations. However, rule acquisition is often the bottleneck that hinders the development of decision systems. When these rules are based on regulations written in Natural Language (NL), one solution is to derive formal business rules from the source documents. This approach also allows changing the rule base with the regulation when it is updated.To address these problems, within the FP7 ONTORULE project, we have designed a method to formalize a base of business rules from NL regulations. We have also developed a tool that supports the work of domain experts and engineers knowledge in this process. This method of formalization involves several steps: a domain expert first extracts the relevant text fragments and rewrites them in the form of independent and autonomous rules written in controlled language; these normalized rules are then formalized for the target application by a knowledge engineer.The talk will present the method of formalization and the tool that has been designed to assist domain experts and knowledge engineers in the exploration of textual sources and to help them interact in the formalization of business rules.

International audience; This paper presents an original use of SBVR to help building a set of business rules out of regulatory documents. The formalization is analyzed as a three-step process, in which SBVR-SE stands in an intermediate position between the Natural Language on the one hand and the formal language on the other hand. The rules are extracted, clarified and simplified at the general regulatory level (expert task) be- fore being refined according to the business application (engineer task). A methodology for these first two steps is described, with different op- erations composing each step. It is illustrated with examples from the literature and from the ONTORULE use cases

Due to the rapid evolution of knowledge areas, the maintenance of the information systems becomes increasingly difficult to manage. In order to ensure the flexibility of such systems, we propose an approach that enables to separate the business domain knowledge from the hard coded application. This solution enables business users to maintain the evolution of the business domain knowledge and facilitates the maintenance of the information systems. The proposed solution consists of integrating ontologies and business rules. Ontologies enable modeling domain knowledge. Based on that domain knowledge, business rules are authored, in a controlled natural language, in order to automate decision. As business rules depend on ontology entities, we focus on the ontology evolution impact on rules. For that, we propose the MDR approach (Model - Detect - Repair), which consists on modeling ontology changes, detecting the problems they may cause and proposing repairs to solve these problems. The MDR approach is business users oriented and is based on Business Rules Management Systems., Vu la rapidité de l'évolution des connaissances des domaines, la maintenance des systèmes d'information est devenue de plus en plus difficile à gérer. Afin d'assurer une flexibilité de ces systèmes, nous proposons une approche qui permet de représenter les connaissances des domaines dans des modèles de représentation des connaissances plutôt que de les coder, dans un langage de programmation informatique, dans l'application du domaine. Ceci assurerait une meilleure flexibilité des systèmes d'information, faciliterait leur maintenance et permettrait aux experts métier de gérer eux même l'évolution des connaissances de leur domaine. Pour cela, nous proposons une approche qui permet d'intégrer des ontolo- gies et des règles métier. Les ontologies permettent de modéliser les connais- sances d'un domaine. Les règles permettent aux experts métier de définir et d'automatiser, dans un langage naturel contrôlé, des décisions du métier en se fondant sur les connaissances représentées dans l'ontologie. Ainsi, les règles dépendent des entités modélisées dans l'ontologie. Vu cette dépendance, il est nécessaire d'étudier l'impact de l'évolution des ontologies sur les règles. Pour cela, nous proposons l'approche MDR (Modéliser - Détecter - Réparer) qui permet de modéliser des changements d'ontologies, de détecter les problèmes de cohérence qu'ils peuvent causer sur les règles métier et de proposer des solutions pour réparer ces problèmes. L'approche proposée est une approche orientée experts métier et est fondée sur les systèmes de gestion des règles métier.

TERMINAE est une plateforme d'aide à la construction de ressources termino-ontologiques à partir de ressources textuelles. Elle met en oeuvre la méthodologie Terminae développée au LIPN. L'une des caractéristiques de cette méthode de construction de ressources termino-ontologiques est la mise en évidence d'un niveau termino-conceptuel constitué par un réseau terminologique normalisé, sorte de passerelle entre le niveau linguistique et le niveau conceptuel. Cette plateforme réalisée par une application eclipse intégre la visualisation de résultats d'outils de traitement automatiques des langues ( extracteurs de termes, outils de reconnaissances d'entités nommées) à des composants permettant de construire des ressources termino-ontologiques (terminologies, thésaurus, ontologies) avec un ancrage linguistique. La RTO construite est couplée à une composante linguistique constituée des éléments textuels ayant servis de base à la création des éléments constituant la ressource. La plateforme supporte l'import/export en SKOS. Dans la version actuelle, le niveau conceptuel est réalisé par le plugin Neon Toolkit 2.4 (http://neon-toolkit.org/wiki/Main_Page). Les langages disponibles sont le français et l'anglais.

International audience; L'extraction automatique de termes est utilisée pour des tâches variées comme l'analyse terminologique, la détection des mots clés pour la recherche d'information et la construction d'ontologies. Les outils de traitement automatique de la langue (TAL) ont la charge d'extraire les termes d'un domaine à partir de corpus spécialisés, mais ces outils n'extraient pas que des termes pertinents. Notre objectif est d'améliorer la sélection des termes pour un domaine donné. Nous proposons des méthodes de filtrage et de pondération de termes qui tiennent compte de la distribution des termes au voisinage des entités nommées et nous montrons qu'elles aident à détecter les termes représentatifs d'un domaine.

International audience; This paper describes a platform that helps industrial domain experts to preserve the connection between textual sources and formalized business rules by using lexicalized ontologies both for links and for storage of the conceptual knowledge. Business Rules Management Systems (BRMSs) are used to update and query business rules of an automotive use case. They rely strongly on domain ontologies, which model the business knowledge and provide a conceptual vocabulary for the formalization of the rules that are expressed in written policies. We show that lexicalized ontologies are a key component of such BRMSs and how such knowledge can be encoded. Our proposed solution supports domain experts in the automotive industry in understanding and maintaining their business rules by presenting the relevant source documents that were used to create the ontological concepts. The use case is based on a car development scenario that models the connection between car testing scenarios, e.g., safety tests, and the methods and tools used to analyze and prepare these tests. The intended solution has been developed in the ONTORULE project and is still work in progress.

International audience; Ontologies that have been built from texts can be associated with lexical information that is crucial for the semantic annotation of texts and all semantic search tasks. However, the entire pocess of building ontologies from texts cannot be fully automated and it is important to guide the knowledge engineer during the building process. This paper presents an enriched version of TERMINAE, which is a text-based methodology for ontology design. It combines a fact-based approach of modeling with the more traditional concept-centric one. We show that named entities can be used to enrich an existing ontology and to bootstrap the acquisition process. In other words, named entities are used for the conceptualisation of ontologies and not only for their population. This approach is illustrated on two use-cases based on policy documents and evaluated by measuring the Precision and Recall of the resulting ontologies with respect to pre-existing ontologies independently built by domain experts.

International audience; This paper describes a simple formalism designed to encode lexicalized ontologies and shows how it is used in a business rule management platform of the automative domain.

International audience; Building ontologies from plain texts is still a research issue. This process cannot be fully automated but natural language processing and methodological guidelines can help the knowledge engineer's task. In this paper we present TERMINAE and show through the analysis of three different experiments on policy documents how the initial terminological approach can be guided by taking named entities into account.

International audience; The contemporary world produces huge bodies of policies and regulations, while the underlying procedures tend to be automated in decision systems, which are designed to define, deploy, execute, monitor and maintain the various rules to which an organization or enterprise has to comply. It is important that the written documentation is integrated into such decision systems in order to refer to the texts to explain decisions, to update the systems when the policy evolves or, conversely, to amend the source documents if some of the rules happen to be inconsistent. The problem is that the complexity of information to be searched for is not reachable by an automated processing, but that their volume prohibits a manual one. Arguing that the integration of policies in decision systems can be better achieved through a semantic annotation than by the full parsing of the source documentation, this paper presents a technical environment that enables the building and exploitation of such semantic annotations.