EMBEDDIA – Cross-Lingual Embeddings for Less-Represented Languages in European News Media
The EMBEDDIA project is developing cross-lingual solutions for text analysis and generation, building foundations for the multilingual next generation internet, for the beneﬁt of European citizens and industry.
By leveraging cross-lingual word embeddings and deep neural networks, we are developing a new generation of language technology tools that are both multilingual (able to deal with or produce content and resources in multiple languages) and cross-lingual (transfer easily across languages), and that can therefore support less-resourced European languages such as Estonian, Croatian and Slovenian.
EMBEDDIA supports the media industry in the age of fast digital media transformation to provide EU citizens better and more personalised access to media content. In particular we focus on solutions for:
- news analysis (e.g. detecting keywords, viewpoints, extracting news summaries)
- comment filtering (e.g. hate speech and troll detection)
- news generation (e.g. supporting automated article creation and creative headline generation).