Wednesday 8th December, 9am to 1pm (CET)
Newsrooms increasingly use and rely on AI tools for automatic text processing. However, these are mostly developed for major languages and that limitation continues to be a challenge. New tools allowing high quality transformations between languages are urgently needed.
In a half-day industry conference that will be held online on December 8, 9am to 1pm (CET), you will have the opportunity to be the first to find out about tools developed within the EMBEDDIA project. EMBEDDIA is a Horizon 2020 funded project consisting of a large European consortium of partners from academia, media and technology. EMBEDDIA has developed tools designed for newsrooms, but also applicable to other text-based sectors based on cross-lingual word embeddings for less-represented languages.
During the conference we will demonstrate the EMBEDDIA Media Assistant which comprises a collection of tools for AI based text processing. Solutions for the following four tasks will be presented through concrete media industry cases: 1) comment filtering and hate speech detection, 2) keyword extraction, 3) article tagging via named entity recognition and linking, and 4) text generation in newsrooms. All tools will be available free of charge and released under open source licences.
There will also be a keynote by Professor Neil Maiden, Chief Product Officer, JECT.AI, about co-designing new technologies for journalists.
As Europe becomes more multicultural, and personal and professional mobility between cultures rapidly increases, access to fundamental resources such as local news and government services is limited by the great diversity of the EU’s 37 languages. The EMBEDDIA project addresses these challenges by leveraging innovations in the use of cross-lingual embeddings coupled with deep neural networks to allow existing monolingual resources to be used across languages, leveraging their high speed of operation for near real-time applications, without the need for large computational resources.
Contact: Carl-Gustav Lindén, University of Helsinki carl-gustav.linden@helsinki.fi
Conference Progamme (all times CET):
09:00 Introduction: AI Technology for the Media Industry: Carl-Gustav Lindén (University of Bergen), Marko Milosavljevic (University of Ljubljana)
09:20 EMBEDDIA project, main results and technologies overview: Senja Pollak (Jožef Stefan Institute)
09:40 Comment filtering and hate speech detection demo: Matthew Purver (Queen Mary University of London), Marko Pranjić (Trikoder)
10:05 Keyword extraction: Senja Pollak, Ivar Krustok (Ekspress Meedia)
10:30 Coffee break
10:40 Text generation in newsrooms: Leo Leppänen (University of Helsinki), Salla Salmela (STT-Lehtikuva)
11:00 EMBEDDIA media assistant and TEXTA toolkit: Silver Traat (Co-founder and COO, Texta)
11:30 Keynote: Co-designing new technologies for journalists to adopt: Experiences from the coalface, Neil Maiden (Professor of Digital Creativity, City University of London & Chief Product Officer, JECT.AI)
12:00 COBAIN: Challenges, Opportunities and Best prActices In News media industry – Round table moderated by Matthew Purver
12:30 Breakout rooms on the practical industrial use of EMBEDDIA comment filtering, keywords, text generation tools
13:00 End of conference
