Cyril Faucher

Evaluation of Kermeta for solving graph-based problems

Kermeta is a meta-language for specifying the structure and behavior of graphs of interconnected objects called models. In this paper, we show that Kermeta is relatively suitable for solving three graph-based problems. First, Kermeta allows the specification of generic model transformations such as refactorings that we apply to different metamodels including Ecore, Java, and Uml. Second, we demonstrate the extensibility of Kermeta to the formal language Alloy using an inter-language model transformation. Kermeta uses Alloy to generate recommendations for completing partially specified models. Third, we show that the Kermeta compiler achieves better execution time and memory performance compared to similar graph-based approaches using a common case study. The three solutions proposed for those graph-based problems and their evaluation with Kermeta according to the criteria of genericity, extensibility, and performance are the main contribution of the paper. Another contribution is the comparison of these solutions with those proposed by other graph-based tools.

Applying Semantic Web Technologies for Improving the Visibility of Tourism Data

Tourism industry is an extremely information-intensive, complex and dynamic activity. It can benefit from semantic Web technologies, due to the significant heterogeneity of information sources and the high volume of on-line data. The management of semantically diverse annotated tourism data is facilitated by ontologies that provide methods and standards, which allow flexibility and more intelligent access to on-line data. This paper provides a description of some of the early results of the Tourinflux project which aims to apply semantic Web technologies to support tourist actors in effectively finding and publishing information on the Web.

Temporal knowledge acquisition and modeling

The objectives of this paper are to present, describe, and explain the foundations and the functionalities of a temporal knowledge acquisition and modeling solution workflow, which aims at acquiring temporal knowledge from texts in order to populate a constrained object model. We are using several models for temporal data, one of which is generic and employed as a pivot model between a linguistic representation and a calendar representation. The approach we propose is generic and has been tested against a real use case, in which input data is made of temporal properties defining when a given location (a theater, a restaurant, a shopping center, etc.) is open or closed. Most expressions entered are expressed in intension. Our models provide a core support to the system that linguistically analyses data entries, transforms them into extensive calendar information and allow users to control the quality of the systems interpretation.

Location-and temporal-based services for nature-society interaction regulation

With the surge of mobile devices and applications, knowledge base browsing and querying facilities must be adapted to new kinds of users and services. Space and time databases are concerned and as such are the subject of this article. Within a mobile context, data retrieval services must take the location of the call into consideration as regards space and time, insofar many pieces of information do depend on ‘where’ and ‘when’ constraints. We present here with a general object model dedicated to specifying time properties. The novel aspects of our proposal consist of three main points. First, the temporal object model stands as a pivot model, and all the more leverages interoperability between software applications, since it extends common standards, such as ISO 19108 and iCalendar. Second, we accompany the object model with a formal grammar close to natural languages, which helps end users in managing and checking temporal object instances. Finally, our proposal focuses on intensional temporal expressions instead of extensional sets of concrete calendar dates. This allows to express semantic aspects – namely for periodic (cyclic) events – which remain hidden and not computable when only a series of dates are given. We evoke two kinds of applications for which our model is beneficial; one concerns services called by human users and the other addresses calls placed by software agents e.g. in a multi-agent system for instance within a simulation context.

CEEPS4IoT: Context-aware Energy Efficient Publish/Subscribe for Internet of Things

Data collection in Internet of Things is greatly impaired by the redundant or unwanted reportings provided by sensors deployed in urban area. The service provider/ application is typically unaware of an objects context while collecting its readings. This paper proposes CEEPS4IoT as a publish- subscribe context-aware system for data collection that takes into account the context of neighboring sensors while collect- ing data. However, rational sensors do not cooperate since sharing their readings is an energy costly operation along incurring them additional communication overhead. To cater this, we present a dynamic coalition game for sensors to collaborate and share their readings in an energy-efficient way and in return receive a reward for cooperating. We derive a stable utility for a sensor proportional to the amount of data it shares while compensating for its energy costs. Results from evaluating CEEPS4IoT in networks of up to 300 nodes suggest it as a scalable and energy efficient pub/sub system with context awareness since it conserves around 50% energy compared to existing pub/sub system.

OAISIS: An ontological-based approach for interlinking CrowdSensing information systems

Nowadays, smartphones and wearable devices are endowed with several sensors, which can harvest large quantities of data about urban areas (location information, pollution levels, etc.), going through a list of personal and surrounding contexts such as noise level, traffic awareness, to name a few. Exploiting this wealth of information provided by the crowd allows developers to design and build several applications over the so-called Mobile CrowdSensing, such as traffic regulation, environmental monitoring, tourism recommendation, etc. However, for this to be possible, many barriers still have to be overcome such as collecting, handling, structuring and representing the crowd data in a suitable way. In this paper, we propose a three-fold solution to the problem of data management in CrowdSensing systems. Firstly, we structure the collected data based on semantic ontologies. Secondly, we enrich the data based on a novel contextual awareness data interlinking. Finally, we refine recommendations with contexts, through taking into consideration meta-information interlinked to the main information of interest. We have implemented our model in a tourism recommendation application as a proof of concept. The experimental evaluation - which we carried out - has shown very promising results.

Une approche fondée sur l’IDM pour le développement d’un environnement de production journalistique

This paper presents an application of Model Driven Engineering to the design of a complete news processing line within the scope of culture and leisure information. MDE is the only technology actually enabling to master the management of complex composite events with spatial and temporal properties, and to cope with making heterogeneous applications interoperate. We provide an overview of our approach and outline the salient points of our metamodels. The stress is put on the event metamodel, standing as a pivot between the beforehand Natural Language Processing, and temporal information visualization frameworks on the end user side. MOTS-CLES : validation semantique, IDM, information temporelle, evenement.