For best experience please turn on javascript and use a modern browser!
You are using a browser that is no longer supported by Microsoft. Please upgrade your browser. The site may not present itself correctly if you continue browsing.
Former Bachelor’s students Roan Schellingerhout (Information Science), Maik Larooij (Information Science) and Danny van den Berg (Artificial Intelligence) have published articles and conference papers based on their thesis. Their supervisors and lecturers are very proud at their achievements.

Roan Schellingerhout

Although Roan already completed his undergraduate studies in Information Science cum laude in 2021, he recently published a paper based on that thesis. His thesis, graded a 9.5 and supervised by Maarten Marx, was titled 'YouTube's Treatment of Conspiracy Content: An Analysis of the You Tube Algorithm'. Based on his thesis, the article 'Accounting for Personalization in Personalization Algorithms: YouTube's Treatment of Conspiracy Content' was published in the journal Digital Journalism. The article ended up on the journal's 'Most Read articles in the Last Year' list.

After having successfully completed his Bachelor’s programme, he then concluded the Faculty’s MSc programme Data Science also cum laude. His Master's thesis ('Explainable Career Path Predictions Using Neural Models') was also recently published, in collaboration with Maarten Marx.

Roan is currently pursuing a PhD at the University of Maastricht.

Maik Larooij

Former BSc Information Science student Maik Larooij turned his BSc thesis into two publications: one in ESB, a Dutch journal in the field of economics (titled 'Open government can improve almost free of charge') and a conference paper in March 2023, at the 45th European Conference on Information Retrieval (ECIR), called ‘Enticing Local Governments to Produce FAIR Freedom of Information Act Dossiers'. His thesis was titled 'The FAIRification of Woo files', for which he was graded a 9.

Maik: "On 1 May 2022, the new Open Government Act (Woo) went into effect. Right around the time I started working on my thesis. In my thesis, I assessed current government publications against the FAIR data principles. Often they did not comply with the new Open Government Act (Woo). Documents are often not machine searchable and lack relevant metadata. With our continued work, we are trying to tackle this problem at the source. Among others, we visited the Ministry of the Interior, several municipalities, the VNG (Association of Dutch Municipalities) and the ACOI (Publicity and Information Management Advisory Board). In our piece for the European Conference on Information Retrieval (ECIR), we write about the process of 'nudging' governing bodies to publish better.

With the publication in ESB in December 2022, we showed that the proposed changes need not cost much extra money at all; something many governing bodies feared. A more recent collaboration is with the ACOI. With them, we hope to attract the attention of the government and parliament. Meanwhile (October 2023), our search engine is well stocked: almost one and a half million pages of Woo documents from many different governing bodies in 10 different categories. We hope to move towards 2 million pages soon. Woogle has also recently switched to an Elasticsearch back-end for even faster and more precise searching. Our primary goal is to show governing bodies the tangible benefits of good publications, exemplified through Woogle."

This year (2023-2024) Maik started the joint UvA-VU Master's in Computer Science.

Danny van den Berg

Former Bachelor's student Artificial Intelligence Danny has been submitting and presenting a conference paper on his BSc thesis to IJCNN (International Joint Conference on Neural Networks), titled ‘Data Under Siege: The Quest for the Optimal Convolutional Autoencoder in Side-Channel Attacks’.

“Our publication researches the robustness of encryption systems against side-channel attacks, a significant security concern. We explored the use of Convolutional Autoencoders (CAE) further in order to enhance these attacks. In doing so we significantly improved attack efficiency by up to 37% against noise and reducing parameters with up to a factor of 29. This work was published at IJCNN which is a conference by IEEE (Institute of Electrical and Electronics Engineers). Continuing exploration of this topic is being taken up by new bachelor students, however I will be considering extending this research into my master’s thesis as well. Currently I am pursuing a Master’s degree in Artificial Intelligence at the Faculty of Science. The feedback from peers has been very positive, and lecturers have shown a keen interest.”