About

Babelfy is a unified, multilingual, graph-based approach to Entity Linking and Word Sense Disambiguation based on a loose identification of candidate meanings coupled with a densest subgraph heuristic which selects high-coherence semantic interpretations.
Babelfy is based on the BabelNet 3.0 multilingual semantic network and jointly performs disambiguation and entity linking in three steps:

  • It associates with each vertex of the BabelNet semantic network, i.e., either concept or named entity, a semantic signature, that is, a set of related vertices. This is a preliminary step which needs to be performed only once, independently of the input text.
  • Given an input text, it extracts all the linkable fragments from this text and, for each of them, lists the possible meanings according to the semantic network.
  • It creates a graph-based semantic interpretation of the whole text by linking the candidate meanings of the extracted fragments using the previously-computed semantic signatures. It then extracts a dense subgraph of this representation and selects the best candidate meaning for each fragment.
The text, written in any of the 271 languages supported by BabelNet 3.0, is output with possibly overlapping semantic annotations. For example, suppose we want to disambiguate the sentence: "Nintendo announces new details on Mario Kart 8.". The application of Babelfy gives the following output:


Moreover, a novel language-agnostic setting is available. In this setting, Babelfy takes into account all 271 languages without assuming or trying to infer the language of the input text giving the further possibility of annotating text written in multiple languages.

For example, we show the output for the following sentence when the language AGNOSTIC is selected: BabelNet is both a dizionario enciclopedico multilingŁe und a reseau semantique








Credits

Concept: Andrea Moro and Roberto Navigli
Programming: Andrea Moro, Federico Scozzafava and Francesco Cecconi
Web application design: OPPOSITE™, Francesco Cecconi and Luca Matteis
Logo: Lorenzo Ceccotti