Projects & Journalism

In the context of SlimZoeken, we used web search as a lens to investigate the following question:

• What is the content that (Dutch) children find on the internet?

For this study, we selected one hundred queries from a list of two hundred random, authentic searches by primary school pupils.

We have manually annotated the search results to evaluate the quality and suitability of web search results for a specific audience. The presented approach

• provides real-world insights into the web as seen through the eyes of Dutch children, and

• introduces a method that can be applied to various cases.

Output:

• Git repository with notebook for the data analysis

• Presentation at the i&i spring conference 2024 (in Dutch; Conference for Computer Science & Digital Literacy Teachers)

• Presentation at the HSN Conference 2024 (in Dutch; Conference for Teaching Dutch)

The Syllabus is a project related to the Center for the Advancement of Infrastructural Imagination (CAII):

We are a non-profit knowledge curation platform committed to defending and strengthening a well-informed public sphere. To do that, we strive to unearth, disseminate, and highlight high-quality information – without deepening public dependence on opaque algorithmic solutions pushed by Big Tech. We do so both by offering individual subscribers a “clean” feed of high-quality content – and by working with institutions (think-tanks, foundations, media, companies) who have their own bespoke information needs.

I have been involved since the early stage of the project, designing and implementing Natural Language Processing (NLP) solutions to semi-automatically support the curation process.