Experience
-
Sep 2020 - Present
PhD Candidate
Radboud University Nijmegen
Behavioural Science Institute
Research on natural language generation and dialogue management for health communication with conversational agents.
Projects: Look Who's Talking
-
Sep 2017 - Feb 2022
Computational Linguist
FloodTags
Development of machine learning-based systems for event detection and information extraction, with a focus on analyzing news articles related to natural disasters. Created an end-to-end information processing pipeline that has been successfully applied and tested in multiple languages, including English, French, Burmese, Lao, and Khmer.
Projects: FloodTags NEAT
-
Mar 2020 - Jul 2020
Scientific Software Developer
Radboud University Nijmegen
Centre for Language and Speech Technologies
Developed a system that converts Dutch dialect words into their dictionary forms. The project also included a demo website where users can interact with the system.
Projects: Dialect2Keyword
-
Apr 2018 - Sep 2019
Guest Researcher
Radboud University Nijmegen
Centre for Language Studies
Research focused on information extraction from unstructured text, including applications of it on low resourced languages.
Projects: FloodTags NEAT
-
Apr 2016 - Aug 2017
Software Developer
FloodTags
Developed a machine learning-based information extraction system to detect critical information about disasters in news articles. Additionally, started a data visualization website which displays the processed data in real time.
Projects: FloodTags NEAT, FloodTags Dashboard
-
Sep 2015 - Mar 2016
Scientific Software Developer
Radboud University Nijmegen
Centre for Language and Speech Technologies
Developed web interfaces and RESTful APIs for several text mining projects.
Projects: Relevancer, Sinfex
-
Feb 2015 - Jun 2015
Teaching Asistant
Muğla Sıtkı Koçman University
Department of Computer Engineering
Taught Linux basics, had practical programming sessions for first year computer engineering bachelor students.
-
Jun 2014 - Sep 2014
Intern
Radboud University Nijmegen
Centre for Language and Speech Technologies
Developed a web interface for a project on event detection from social media on the basis of two PhD studies
Projects: Lama Events
-
Feb 2013 - Jun 2014
Android Application Developer
Muğla Sıtkı Koçman University
Department of Computer Engineering
Developed the official Android application of the university, alongside minor web applications for the computer engineering department.
Projects: MSKU Mobil
Education
Artificial Intelligence
Radboud University Nijmegen
Master of Science
2015-2017
Projects
Look Who's Talking
Developing long-term engaging chatbots to conduct human-like conversations with humans for two technologically and linguistically diverse domains; smoking cessation and safe sex promotion.
Kon ta bAI?
Developing a large language model that is able to understand and generate text in the low-resourced Papiamentu language, which has about 350.000 native speakers in the world.
FloodTags NEAT
Newspaper Event Analysis Tool is a machine learning based end-to-end information processing pipeline performs event detection and information extraction to analyse news articles and collect information about natural disasters.
Dialect2Keyword
A Natural Language Processing application and a web interface to convert words in Dutch dialects into standard Dutch keywords which can be used as the basis for interoperable searches through various dialect dictionaries.
FloodTags Dashboard
An online dashboard to monitor online media for flood information. The Dashboard shows what people are saying about floods in the media. This enables online monitoring new developments related to floods.
Relevancer
Relevancer aims at finding a balance between complexity of a tweet data set and expert needs from this data. The process can be summarized as creating a need based classifier without sacrificing precision, information, and time.
Lama Events
Lama Events is a calendar application listing events in the near future. The events are detected and selected by a fully automatic procedure in the Dutch Twitter stream.