Research projects

("DISCO-SUMO"; NWO Creative Industries programme; 2015-2018)

Summary:
People access the internet increasingly through mobile devices and simultaneously spend more time reading social media than traditional media.

One of the important functions of discussion and review websites is to facilitate social interaction in the form of discussions. However, the structure of the discussions and reviews posted on these websites is difficult to render on mobile devices. They can be long, and the highlights can be hidden anywhere in the thread. A small screen is a serious hindrance to searching and browsing in this mass of text.

To address this problem a team of researchers from Radboud University, Tilburg University, and Sanoma Media BV will develop a new discussion thread summarizer, which will allow mobile users to quickly digest the key points of a discussion thread as it has developed so far, thereby making it easier for users to contribute. As a result, the threshold for mobile users to participate in online discussions will be significantly lowered, and the general quality and understanding of these discussions may be expected to increase.

Website: http://discosumo.ruhosting.nl/

Twitter: https://twitter.com/disco_sumo

the Case of the Hadza Hunter-Gatherers of Tanzania

(Marie Skłodowska-Curie Individual Fellowship; 2015-2017)

The way children learn to communicate verbally and non-verbally depends to a great extent on the culture in which they are raised. Little is known about children's communicative development in hunter-gatherer societies.

The objective of the proposed project is to gain insights into this in Hadza hunter-gatherers in Tanzania. A combination of methods will be used, among others a measuring device that records multiple modalities, including heart rate and audio recordings.

("COSLI"; NWO-Artificial Intelligence; 2015-2018)

Humans learn to communicate through social and linguistic interaction. When developing artificial agents whose purpose is to communicate with humans or with other agents, it is important to use social cues to enhance the natural quality of communication. To date, artificial intelligence models of communicative agents either have no social interaction component, or their behavior tends to follow fixed and pre-specified patterns that does not resemble human interaction behavior.

In this project, we aim to develop an agent-based model that simulates naturalistic social and linguistic interactions in order to learn and use language for communication purposes.

("APOLOGY"; ERC-Consolidator Grant; 2016-2021)

Summary:
In the past decades, there has been a considerable rise in the number of apologies offered by states for injustices and human rights violations. At present, however, we do not know whether political apologies are a universally viable way to restore justice and harmony.

This project addresses this challenge. Using an innovative, interdisciplinary, and multi-method approach with in-depth interviews, (experimental) surveys, and content analyses of apologies, it will be analyzed whether there are universals in how political apologies are valued, expressed, and interpreted or whether this varies as a function of cross-cultural differences in key values (collectivism and individualism) and norms (face and honor). Based on the findings, we will build a theoretical framework regarding the potential value and role of apologies in transitional justice processes.

(NWO Free Competition Humanities; 2015-2019)

Summary:
This projects aims to investigate (1) how the emotional states of speakers influence the language they produce and (2) how the influence of emotion on language production can be modeled in computational tools for affective natural language generation.

Specifically, we ask both whether content selection ("deciding what to say") and message formulation ("deciding how to say it") are affected by emotional appraisals of the language user. This will be studied in a series of experiments, zooming in on referential communication. The experimental findings will feed into the development of a novel automatic news report generation tool that can adjust the message depending on the emotion it is intended to reflect.

("L2TOR"; H2020; 2016-2018)

Summary
L2TOR (pronounced ‘el tutor’) is a scientific research project, called Second Language Tutoring using Social Robots, funded by the Horizon 2020 programme of the European Commission. The project aims to design a child-friendly tutor robot that can be used to support teaching preschool children a second language (L2) by interacting with children in their social and referential world.

Project website: http://www.l2tor.eu/

Twitter: https://twitter.com/L2Tor

("TraMOOC"; H2020; 2015-2018)

Summary
TraMOOC is a H2020 project aiming at providing reliable machine translation for Massive Open Online Courses (MOOCs). The main expected outcome of the project is a high quality machine translation service for all types of educational textual data available on a MOOC platform. The service will support 11 target languages.

The core of the service will be open-source, with some premium add-on services which will be commercialised. Open source will turn the MOOC translation service into a platform that will enable the integration of any machine translation (MT) solution in the educational domain, for any language.

Website: http://tramooc.eu/