The second Maria de Maeztu Strategic Research Program (CEX2021-001195-M) of the Department of Information and Communication Technologies (DTIC) takes place between 2023 and 2026. The website for this program is under construction. You can find some details in this news.

The first María de Maeztu Strategic Research Program (MDM-2015-0502) took place between January 2016 and June 2020. It was focused on data-driven knowledge extraction, boosting synergistic research initiatives across our different research areas.

Back Francesco Ronzano wins the 1st Language Technologies Hackathon at 4YFN

Francesco Ronzano wins the 1st Language Technologies Hackathon at 4YFN

‘TextDigester’, an application that automatically generates textual summaries presented by Francesco Ronzano (Natural Language Technologies Group at DTIC), has won the ‘I Language Technologies Hackathon’. The final took place in the context of the ‘4 Years From Now’ event. Francesco received the award from the Secretary of State for the Information Society and Digital Agenda, José María Lasalle.

28.02.2017

 

The application ‘TextDigester’ presented by Francesco Ronzano (Natural Language Technologies Group at DTIC) won the I Language Technologies Hackathon. The final phase of the hackathon took place in the context of the 4 Years From Now’ event. Francesco received the award from the Secretary of State for the Information Society and Digital Agenda, José María Lasalle.

 

‘TextDigester’ integrates several Natural Language Processing tools to support the creation of summaries of single and multiple documents in different languages. ‘TextDigester’ can retrieve and analyse textual contents from HTML pages (such as a newspaper), XML documents (such as RSS feeds) and JSON objects (such as tweets). Once the language is detected, ‘TextDigester’ performs lexical and semantic analyses of the documents, in order to select and combine the excerpts that better summarize their content. ‘TextDigester’ enables its users to choose among several summarization approaches that can be easily composed and customized. ‘TextDigester’ is implemented as a self contained Java library that integrates NLP frameworks like Freeling, GATE, Deeplearing4j and SUMMA. The code of the application is publicly available on GitHub.

 

The winner was chosen by means of two different votes, both with the same weight. The first one was expressed by the participants of the Hackathon, and the second one by the Jury, formed by María Fernández Rancaño (Red.es), Horacio Rodríguez Hontoria (Universitat Politècnica de Catalunya), Lluís Padró Cirera (Universitat Politècnica de Catalunya), Núria Bel Rafecas (Universitat Pompeu Fabra) y David Pérez Fernández (SESIAD). It is to highlight that both votes were completely coincident with respect to the first 3 positions.

 

The other finalists covered a wide range of topics including chatbots, tools to support learning and information retrieval, presented by individuals, companies such as IBM and Everis, and research centers such as Vicomtech.

 

Additional material_

- Slides used for the presentation (in Spanish)

- Text Digester in Github

Multimedia

Categories:

SDG - Sustainable Development Goals:

Els ODS a la UPF

Contact

Department of Information and Communication Technologies, UPF

Grant CEX2021-001195-M funded by MCIN/AEI /10.13039/501100011033


 

Department of Information and Communication Technologies, UPF

Tànger building (Poblenou campus)
Tànger, 122-140
08018 Barcelona

[email protected]

  • Àngel Lozano - Scientific director
  • Aurelio Ruiz - Program management