Natural Language Processing Research and Development at the Joint Research Center: An Overview of the EMM Linguistic Processing Chain and Related Projects

aula Co-Working / Co-Working room
Palazzo di Lingue

Emmanuel Cartier (Joint Research Center, T5)

^{Natural Language Processing Research and Development at the Joint Research Center: An Overview of the EMM Linguistic Processing Chain and Related Projects}

This seminar will present the various activities carried out in Natural Language Processing at the European Commission’s Joint Research Centre, in the Text and Data Mining unit (T5), where various specialists cooperate (computational linguistics scientists and developers, data analysts, end users) to offer Natural Language Processing tools at the service of the Commission's policy.
The focus will be made on the European Media Monitor (EMM) linguistic processing chain, comprised of several components enabling to enrich the raw data (web scraping, categorisation of web pages, language detection, keyword extraction, Named Entity Recognition and Linking, Sentiment and Emotion Analysis, Quotation extraction, Event extraction, Translation, etc.). The current state and the ongoing developments will be presented. Applied on-going Projects inspired and/or based on this huge multilingual repository will also be presented, like the Epidemic Intelligence from Open Sources (EIOS).

Bibliografia / Select bibliography:
Ehrmann, M., Jacquet, G. and Steinberger, R. (2016), “Jrc-names: Multilingual entity name variants and titles as linked data”, Semantic Web, Vol. 8, pp. 283–295. URL https://api.semanticscholar.org/
CorpusID:2057414.
Steinberger, R. et al. (2017). “EMM: Supporting the Analyst by Turning Multilingual Text into Structured Data.”