Ruben van Heusden

PhD student at the ILPS research group - University of Amsterdam


Ruben van Heusden

My name is Ruben van Heusden and I am a PhD student at the ILPS research group of the University of Amsterdam.
I am currently involved in the ACCESS project and the Second Phase of the ParlaMint project of the CLARIN foundation.

Through both my Bachelor and Master in Artificial Intelligence I have had the opportunity to familiarize myself with a multitude of
concepts and techniques in the exciting field of Artifical Intelligence and Machine Learning. I picked up a particular interest for
everything related to Natural Language Processing and I am very happy to be given the opportunity to conduct more research on NLP at the
ILPS research group.

Current Work

As a PhD student I am working on the ACCESS project. ACCESS stands for: Access to City Councils using Exploratory Search Systems.
The aim of this project is to develop intelligent systems with the use of Artificial Intelligence to aid citizens in more easily accessing and understanding information from different government agencies.

In Collaboration with the Instituut voor Nederlands Taal (INT), the UvA is participating in the Second Phase of the ParlaMint project from ClARIN.
This project aims at making the parliamentary documents from different European countries available in a unified format for easier research.

Past Work

In collaboration with the Gemeente of Amsterdam I have developed and evaluated several Deep Learning methods for Multitask Text Classification. The model uses a Multigate Mixture-of-Experts model and can is tested on different types of text such as emails and written conversation. The thesis can be found on the UvA Scriptie Library



In February - March 2021 I will be working as a Teaching Assistent for the Bachelor Informatiekunde course Data science. This course focusses on making students familiar with the theoretical knowledge and pratical tools essential for Data Science, such as performing data analysis and cleaning data.

Contact Details