source photo : http://2013.eswc-conferences.org

Le LIRIS et l'application DataConf obtiennent la deuxième place du AI Mashup Challenge de la conférence ESWC'2013

The DataConf application received second best submission award in the AI Mashup Challenge, at the ESWC'2013 conference wich was held form 28 to 30 may in Montpellier, France. This application is the result of several research work projects in the Web and Semantic Web fields, carried out by a LIRIS researcher and computer science master degree students of the Université Claude Bernard Lyon 1 (UCBL).

The second place of the DataConf application recompenses the efforts of Florian Bacle, Benoît Durand de la Pastellière, Fiona Le Peutrec and Nicolas Armando, first year students in the UCBL computer science master degree, and Lionel Médini, researcher at LIRIS, in the AI mashup challenge of the European Semantic Web conference (ESWC'2013). This second place is even more meritorious since this high level conference is not accustomed to count master students among its participants, but rather PhD students and confirmed researchers. Competition at the AI Mashup Challenge has also been high-level, since first place was attributed to an application built by doctoral students of the Eurecom research center and SAP company.

DataConf is a Web application that provides conference participants with access to the metadata of the conference using their smartphone, tablet or computer. It allows them accessing the program of this conference, navigating between different information including publications, authors and organizations to which they belong and enriching this data into Linked Data. They can also enrich these contents by providing links to resources giving more information on their own profiles or work.

Technically, DataConf is a Web "mashup" application that aggregates various data sources available on the Web, including knowledge bases (aka SPARQL endpoints) and search engines, as well as sources specifically developed by these students: two Web services dedicated to event management and user-based metadata enrichment, as well as an inference engine embedded in the browser, able to recommend publications to the user, on the basis of his/her past navigation. DataConf uses advanced Web technologies (LocalStorage, Web workers, getUserMedia API), is fully developed to run on client side and therefore requires no server-side processing. It is also configurable and thus can be used for any conference which metadata available on the Web. Moreover, its component-oriented architecture allows easily adding or removing linked data sources.

DataConf is originated from the ‘WWW '2012 Posters’ application, developed by another master degree student (Nguyen Hoang Duy Tan), for the WWW'2012 conference, which was held in Lyon last year in April. It has however been totally restructured and redeveloped by these four students, during a research project of the first year of the UCBL computer science master degree, that aims to introduce students to research work.

The success of this application both from evaluators in the challenge of the participants and the organizers of the conference opened interesting perspectives, both for extending its functionality and using it as a research experimentation platform. The four students, who followed with interest the scientific presentations and established many contacts at the conference, are already thinking of a new version of the application and target a new challenge...

Links:

Access to the DataConf application instances: http://dataconf.liris.cnrs.fr

Documentation (wiki) : http://liris.cnrs.fr/dataconf/