Rating is available when the video has been rented.
This feature is not available right now. Please try again later.
Published on Jun 23, 2014
The creation of huge, broad-coverage knowledge bases and linguistic linked data such as DBpedia and BabelNet, made possible by the availability of collaboratively-curated online resources such as Wikipedia and Wiktionary, has not only entailed researchers, but it has also hit big industry players such as Google and IBM, who are moving fast towards large-scale knowledge-oriented systems. The semantic annotation of arbitrary Web pages is an important use of multilingual linguistic linked data. However, most of the time, the task is limited to just linking (some) named entities or working on a restricted number of languages. In this talk we will present Babelfy (http://babelfy.org), a new, state-of-the-art, wide-coverage system which leverages a novel graph-based algorithm to semantically annotate and link arbitrary text, such as Web pages, written in any language, with both concepts, i.e. abstract meanings of words or domain terms, and named entities from BabelNet (http://babelnet.org), a huge multilingual semantic network and linked data set.