The objective of ADNEXT (ADaptive informatioN EXtraction over Time) is to develop trainable, adaptable Dutch language information extraction technology for named entity recognition, event detection, and time identification. The technology has a broad coverage “default” mode and retrains dynamically to new domains upon being confronted with new (clusters of) news or user-generated data (such as Twitter).
Research into the modelling of source-side context in Machine Translation
Dreams, the involuntary perceptions that occur in our minds during sleep, have been the topic of studies in many fields of research, including psychiatry, psychology, neurobiology, and religious studies. Their narrative content also links dreams to other forms of storytelling, with sharp distinctions (such as the focus on one's personal life and the typical personal perspective) but also interesting overlaps with genres such as orally transmitted folktales. We present a study on dreams aimed at the large-scale analysis of dreams using text analytics.
The FutureTDM project identifies current barriers through policy analysis and consultation with researchers, developers, publishers, and SMEs and will come up with solid European-wide recommendations that address and reduce the barriers on a legal, policy and organizational level.
Integrated Social History Environment for Research: Digging into Social Unrest, a Digging Into Data project
TraMOOC (Translation for Massive Open Online Courses) is a Horizon 2020 collaborative project aiming at providing reliable machine Translation for Massive Open Online Courses (MOOCs). The main result of the project will be an online translation platform, which will utilize a wide set of linguistic infrastructure tools and resources in order to provide accurate and coherent translation to its end users.
In the Vox Populi project we try to monitor the opinion of the people by scanning social media (primarily Twitter). We are interested in political views and especially in predicting the outcome of elections on the basis of tweets. A preliminary study showed that by just counting tweets in which political parties are mentioned, we can approximate the polls. Because the posters on Twitter are not necessarily a cross-section of the population, we will research a number of filters to adjust this possible discrepancy. The filters we will be looking into are about: Clustering and topic detection (what are posts about?), User profiling (who is posting?), Sentiment analysis (what is the sentiment of messages?), External event detection (are there specific events that influence the people’s posts?), Time windowing (should we put relative weights on counts of tweets near a certain event (election)?)