Search

International audience; The training curriculum for medical doctors requires the intensive and rapid assimilation of a lot of knowledge. To help medical students optimize their learning path, the SIDES 3.0 national French project aims to extend an existing platform with intelligent learning services. This platform contains a large number of annotated learning resources, from training and evaluation questions to students' learning traces, available as an RDF knowledge graph. In order for the platform to provide personalized learning services, the knowledge and skills progressively acquired by students on each subject should be taken into account when choosing the training and evaluation questions to be presented to them, in the form of customized quizzes. To achieve such recommendation , a first step lies in the ability to predict the outcome of students when answering questions (success or failure). With this objective in mind, in this paper we propose a model of the students' learning on the SIDES platform, able to make such predictions. The model extends a state-of-the-art approach to fit the specificity of medical data, and to take into account additional knowledge extracted from the OntoSIDES knowledge graph in the form of graph embeddings. Through an evaluation based on learning traces for pediatrics and cardiovascular specialties, we show that considering the vector representations of answers, questions and students nodes substantially improves the prediction results compared to baseline models.

Data published on the web following the principles of linked data has resulted in a global data space called the Web of Data. These principles led to semantically interlink and connect different resources at data level regardless their structure, authoring, location, etc. The tremendous and continuous growth of the Web of Data also implies that now it is more likely to find resources that describe real-life concepts. However, discovering and recommending relevant related resources is still an open research area. This chapter studies recommender systems that use linked data as a source containing a significant amount of available resources and their relationships useful to produce recommendations. Furthermore, it also presents a framework to deploy and execute state-of-the-art algorithms for linked data that have been re-implemented to measure and benchmark them in different application domains and without being bound to a unique dataset.

Data published on the Web following the Linked Data principles has resulted in a global data space called the Web of Data. These principles led to semantically interlink and connect different resources at data level regardless their structure, authoring, location, etc. The tremendous and continuous growth of the Web of Data also implies that now it is more likely to find resources that describe real-life concepts. However, discovering and recommending relevant related resources is still an open research area. This chapter studies recommender systems that use Linked Data as a source containing a significant amount of available resources and their relationships useful to produce recommendations. Furthermore, it also presents a framework to deploy and execute state-of-the-art algorithms for Linked Data that have been re-implemented to measure and benchmark them in different application domains and without being bound to a unique dataset.

International audience; The production of Educational quizzes is a time-consuming task that can be automated by taking advantage of existing knowledge bases available on the Web of Linked Open Data (LOD) such as DBpedia. For these quizzes to be useful to learners, their questions must be generated in compliance with the knowledge and skills necessary to master for each subject and school year according to the official educational standards. We present an approach that exploits structured knowledge bases that contain the knowledge and skills, from which it selects a set of DBpedia resources relevant to a specific school subject and year. This set of resources is enriched with additional related resources, through the NEFCE heuristic. Finally, question generation strategies are applied to the graph generated with this set of resulting resources. Likewise, we provide an evaluation of two knowledge bases and of the proposed NEFCE heuristics.

International audience; Educational quizzes are useful not only to evaluate or test the knowledge acquired by a learner, but also to help her/him to deepen knowledge about a specific domain or topic in an informal and entertaining way. The production of quizzes is a time-consuming task that can be automated by taking advantage of existing knowledge bases available on the Web of Linked Open Data (LOD). However, automatically extracting from the LOD a knowledge graph composed by the information of a set of resources which are relevant to a given specific domain or topic, is a crucial phase for the automatic generation of quizzes. To address this issue, we propose a heuristic that extracts from DBpedia a set of resources related to a given specific domain. Such heuristic has been implemented and used for the automatic generation of quizzes in the geography and privacy domains. We report a comparative user evaluation of it.

Recommender systems RS are software tools that use analytic technologies to suggest different items of interest to an end user. Linked Data is a set of best practices for publishing and connecting structured data on the Web. This paper presents a systematic literature review to summarize the state of the art in RS that use structured data published as Linked Data for providing recommendations of items from diverse domains. It considers the most relevant research problems addressed and classifies RS according to how Linked Data have been used to provide recommendations. Furthermore, it analyzes contributions, limitations, application domains, evaluation techniques, and directions proposed for future research. We found that there are still many open challenges with regard to RS based on Linked Data in order to be efficient for real applications. The main ones are personalization of recommendations, use of more datasets considering the heterogeneity introduced, creation of new hybrid RS for adding information, definition of more advanced similarity measures that take into account the large amount of data in Linked Data datasets, and implementation of testbeds to study evaluation techniques and to assess the accuracy scalability and computational complexity of RS. Copyright © 2015 John Wiley & Sons, Ltd.

International audience; The increase in the amount of structured data published on the Web using the principles of Linked Data means that now it is more likely to find resources on the Web of Data that represent real life concepts. Discovering and recommending resources on the Web of Data related to a given resource is still an open research area. This work presents a framework to deploy and execute Linked Data based recommendation algorithms to measure their accuracy and performance in different contexts. Moreover, application developers can use this framework as the main component for recommendation in various domains. Finally, this paper describes a new recommendation algorithm that adapts its behavior dynamically based on the features of the Linked Data dataset used. The results of a user study show that the algorithm proposed in this paper has better accuracy and novelty than other state- of-the-art algorithms for Linked Data.

International audience; Educational quizzes are very valuable resources to test or evaluate the knowledge acquired by learners and to support lifelong learning on various topics or subjects, in an informal and entertaining way. The production of quizzes is a very time-consuming task and its automation is thus a real challenge in e-Education. In this paper, we address the research question of how to automate the generation of quizzes by taking advantage of existing knowledge sources available on the Web. We propose an approach that allows learners to take advantage of the knowledge captured in domain ontologies available on the Web and to discover or acquire a more in-depth knowledge of a specific domain by solving educational quizzes automatically generated from an ontology modelling the domain. The implementation and experimentation of our approach is presented through the use case of a world-famous French game of manually generated multiple-choice questions.

International audience; In the education field, in order to achieve learning goals, it is necessary to define learning paths that foresee a gradual and incre-mental acquisition of certain knowledge and skills that students should acquire. In this paper we analyze the educational progressions of the French educational system, we show how to formalize them through a web ontology and how to perform knowledge extraction from the official texts describing them to automate the population of such an ontology.

The Web's evolution into a Semantic Web and the continuous increase in the amount of data published as linked data open up new opportunities for annotation and categorization systems to reuse these data as semantic knowledge bases. Accordingly, information extraction systems use linked data to exploit semantic knowledge bases, which can be interconnected and structured to increase the precision and recall of annotation and categorization mechanisms. TellMeFirst classifies and enriches textual documents written in English and Italian. Although various works present solutions for text annotation and classification, this article describes and studies the use case of a telecommunications operator that has adopted TellMeFirst to add value to two services available to its users: FriendTV and Society.

Nowadays the IT Clouds are used to provide services and run applications (in a flexible and scalable way) where the entire lifecycle of the application or service is under the control of the Cloud Management Subsystem that is able to automatically and autonomously start, stop, restart a server or an application. Additionally, it is able to determine when a certain application or service is running under overload conditions and reduce the load by adding additional virtual machines hosting the affected software or releasing the resources not used anymore. These are classic operations performed by nowadays Clouds. However little effort has been dedicated to support a developer during the application development process or during the service execution by changing the application behavior dependently on the service's components response and availability. The Cloud Wave project founded by the European Commission within the FP7 Research Program aims to contribute on these underdeveloped concepts. It contributes to the development process of a service or application in a feedback-driven way and also to the service execution by adapting the service accordingly to the execution environment and condition changes.

The expected evolution of the IoT into a huge and growing number of sensors connected to the Internet and generating a huge amount of data, translates into new challenges in the analysis phase for the extraction of useful information to the final user. This scenario opens new opportunities for the introduction of context-awareness features, as it is possible to store context information linked to the data obtained by sensors to make their interpretation simpler and useful for the user. The essential objective of the EU funded FI-WARE project, is to contribute in the development and implementation of Future Internet services and applications by using and reusing a series of building blocks called Generic Enablers (GE). This paper presents the contribution of the FI-WARE project to the IoT: it makes available to the creators of applications and services, a GE for context management and an IoT Services Enablement Architecture which allows them to create modular applications that promote the reuse of software (reducing the development time and increasing software quality) and are based on standards.

Internet has evolved and grown tremendously during the last years. We're living the rise of the “self-made” Web Services. On the one hand, this conception reduces the limitations for the creation of new Web Services capable of operating on top of the large Internet technological foundation, while on the other it has some drawbacks such as: slow creation process, long time-to-market, deep knowledge of many Internet Technologies, no standardization, no interoperability and no quality control. All the aforementioned characteristics lead to a strong fragmentation of technology and Web Services. The Future Internet initiative instantiation by European Commission in form of Private-Public Partnership (PPP) Program has the objective to populate Internet with common conceptual components (Generic Enablers or GE) to enable a faster and interoperable service creation. FI-WARE is the technological core foundation project of the PPP program that provides specifications and reference implementations of the most common generic enablers and deploys them on its test-bed for public and evaluation usage. The Publish/Subscribe and the Semantic Annotation GEs that the project has provided and installed in the FI-WARE test-bed are demonstrated in this paper.

In this paper, we present an approach to exploit Linked Data in Smart Spaces, doing more than just using RDF to represent information. In particular, we rely on knowledge stored in DBpedia, a dataset in the Web of Data. We also provide a platform to implement such an approach and a eTourism use case, both developed in collaboration with a mobile operator. Finally, we provide also a performance evaluation of the main component of the platform.

International audience; Access control is a recognized open issue when interacting with RDF using HTTP methods. In literature, authentication and authorization mechanisms either introduce undesired complexity such as SPARQL and ad-hoc policy languages, or rely on basic access control lists, thus resulting in limited policy expressiveness. In this paper we show how the Shi3ld attribute-based authorization framework for SPARQL endpoints has been progressively converted to protect HTTP operations on RDF. We proceed by steps: we start by supporting the SPARQL 1.1 Graph Store Protocol, and we shift towards a SPARQL-less solution for the Linked Data Platform. We demonstrate that the resulting authoriza- tion framework provides the same functionalities of its SPARQL-based counterpart, including the adoption of Semantic Web languages only.

Emergency Telecommunications EMTEL deal with provisioning of telecommunications services in emergency situations; these can range from an individual in a state of personal emergency with need to make an emergency call due to sudden illness, traffic accident, fire to a very broad perspective of serious disruptions earthquakes, floods, terrorist attacks. EMTEL covers the telecommunications needs of public safety resources, like: police, fire-fighters, ambulances and other medical services: such services usually rely on dedicated networks and equipment. We realised PICO control platforms and delivery of services in converged networks a platform for service delivery and emergency management on public telecommunications networks. PICO platform leverages on application streaming on Android OS and a context-aware system for fixed-mobile convergence based on IP Multimedia Subsystem IMS.

Nowadays, the variety of mobile services is growing together with the number of customers and the heterogeneity of their mobile devices, thus, monitoring the users of a mobile network has become a challenge. Considering Smart Space Governing to address this issue, in which the mobile network is considered an Smart Space (due to its size and complexity) and their elements (mobile devices and network monitoring services), it is possible to create rules to monitor the output of these elements. We call those rules as Smart Space Compositions and can be created through the platform’s Visual Editor, that during the graphical creation process, provides the user with a list of similar existing compositions that can be reused at any time to improve the composition process. This paper describes the implementation by a Telecommunications Operator of this composition module supported by subgraph isomorphism techniques.

The advent of contemporary mobile devices and their increasing computing power and location capabilities combined with the most innovative Web technologies has provided mobile users with new possibilities to share experiences on-the-go. The growing quantity of multimedia content present on the Web makes it difficult for mobile users to retrieve suitable content. Typically, users looking for interesting content related to their current position or POI (point of interest), access Web search engines relying on keywords to describe their ideas. Unfortunately such descriptions are often subjective and thus retrieval can be ineffective. To address these issues, our platform provides users with an application targeted for modern mobile devices that allows content acquisition and publication. Published content is automatically analyzed and stored on our server with semantic annotations based on the user's context and content, for further semantic search. We describe how and why we migrated from a triple-tags technology to semantics, hoping for related Linked Data.

The rapid and constant evolution of Smart Spaces has introduced an increase in the complexity and quantity of its components, making the monitoring and management processes more challenging and complex. On the other hand, Service Mashups enable the design and development novel and modern Web applications based on easy-to-accomplish end-user service compositions.

We describe two mashups based on context data and multimedia content retrieved from a client application running on mobile devices. This information is processed by our Context Aware Platform that provides specific APIs to deliver context data to our mashups. First of all, these APIs enable context tagging of pictures and videos acquired by the users. Machine tags are adopted in order to enable context-filtered representation of pictures and videos on our content management system. Moreover, platform APIs are used to gather context information that is analyzed by ad-hoc clustering algorithms to detect the users' daily actions. We therefore automatically generate structured blog posts that describe the users' day. The same information is represented as video blogs, too.