Events in June 2018
-
Perception, Cognition and Interaction: Part VI Perception, Cognition and Interaction: Part VI
–
June 11, 2018Talk 1
Speaker
Olivier Corby (WIMMICS)
Title
An Introduction to the Semantic WebAbstract
We present the basic principles of the Semantic Web of Linked Data: an extension of the Web that enables users to define a semantic network at the scale of the Web and exchange data and knowledge on the Web. We introduce the core languages designed by the World Wide Web Consortium: RDF, RDFS, SPARQL. We give some examples with the french DBpedia knowledge base, extracted from Wikipedia, that is managed by the Wimmics team. To finish, we present some of our research topics.Talk 2
Speaker
Amine Djebri (WIMMICS)Title
UncertaintyAbstract
The Web is an open world for information, where everyone can say whatever they want. However, when it comes to the Semantic Web, the rules of the game must be changed. The assumption of an open world brings into light new terms such as: vagueness, ambiguity and uncertainty. In this talk, we will discuss uncertainty in the general context of information theory, and to what extent the integration of such a dimension in the Semantic Web have effects on logical reasoning process. -
Perception, Cognition and Interaction: Part VII Perception, Cognition and Interaction: Part VII
–
June 25, 2018Talk 1
Speaker
Oscar Rodriguez-Rocha (WIMMICS)Title
Generating QA Quizzes from DBPediaAbstract
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. However, their creation is a time-consuming task that may also require the knowledge of domain experts. In this context, an automatic approach to the generation of quizzes that takes advantage of existing knowledge bases available on the Web of Linked Open Data (LOD) will be presented. Additionally, through specific use cases, some related research problems and their proposed solutions will be also detailed.Talk 2
Speaker
Raphael Gazzotti (WIMMICS)Title
Preprocessing in Natural Language ProcessingAbstract
On many Natural Language Processing applications where real-world data are involved, it is a necessary thing to proceed to some preprocessing steps in order to achieve the best possible performances -or at least some usable results-. We will review the key principles and methodologies to achieve this goal during this introductory course. Nowadays, it is fairly common to work with heterogeneous data, so if you know how to curate your textual information, it is within your reach to build an efficient system without necessarily depending on the latest most popular API.