MultiJEDI is a 5-year ERC Starting Grant (2011-2016) headed by Prof. Roberto Navigli at the Linguistic Computing Laboratory of the Sapienza University of Rome. The project has two main objectives: creating large-scale lexical resources for dozens of languages, and enabling multilingual text understanding. The project has received funding from the European Union's specific programme ‘Ideas’ implementing the seventh framework programme (FP7-IDEAS-ERC) under grant agreement no. 259234.
The research work follows many directions each addressing one of the project's goals:
The largest multilingual semantic network that integrates WordNet, Wikipedia, Wiktionary and many other resources. BabelNet covers about 14 million concepts and named entities and 271 languages for all open-class parts of speech.
A state-of-the-art graph-based algorithm for joint Word Sense Disambiguation and Entity Linking in arbitrary languages (including a language-agnostic setting!). It exploits the power of the BabelNet multilingual semantic network.
An approach for large-scale knowledge-based harvesting of Semantic Predicates from Wikipedia.
A graph-based approach to learning domain taxonomies from scratch without drawing on existing knowledge resources.
A semantic network obtained by means of an Open Information Extraction approach which extracts semantic relations from Wikipedia sentences and ontologizes them.
An approach to the disambiguation of domain texts using an inventory of senses obtained by means of automated glossary extraction.
An approach for integrating the output of different Open Information Extraction systems into a single unified and fully disambiguated knowledge repository.
A Neural Network approach to encoding the latent meanings of concepts and named entities via an innovative concept of sense embeddings.
A state-of-the-art approach to the explicit semantic representations of WordNet synsets, Wikipedia pages and BabelNet synsets.
A state-of-the-art knowledge-based approach which leverages Personalized PageRank to calculate semantic representations of concepts.
WiBi (the Wikipedia Bitaxonomy) is an algorithm for the automatic creation of a bitaxonomy for Wikipedia pages and categories.
A platform for hosting video games with a purpose for linguistic annotation and validation.
Roberto Navigli is a professor at "La Sapienza" University of Rome. Principal Investigator of MultiJEDI.
A Ph.D. student in graph-based word sense induction and its application to web search result clustering.
A post-doc focused on encoding BabelNet in RDF format and deploying it on SPARQL.
Ph.D. student in the Department of Computer Science at the Sapienza University of Rome.
Ph.D. student in the Department of Computer Science at the Sapienza University of Rome.
Ph.D. student in the Department of Computer Science at the Sapienza University of Rome.
A post-doc focused on video games with a purpose for NLP annotation and validation.
Ph.D. student in the Department of Computer Science at the Sapienza University of Rome.
A research fellow working on several aspects of the BabelNet project and at (LCL).
Ph.D. student in the Department of Computer Science at the Sapienza University of Rome.
Ph.D. student in the Department of Computer Science at the Sapienza University of Rome.
A postdoc working on the first part of the BabelNet project, University of Mannheim.
Ph.D. student in the Department of Computer Science at the Sapienza University of Rome.
Ph.D. student in the Department of Computer Science at the Sapienza University of Rome.
Ph.D. student in the Department of Computer Science at the Sapienza University of Rome.
A Ph.D. student working on word sense induction at the Sapienza University of Rome.