I like working on information filtering challenges in large multimodal archives. My current research interests fall under two categories, which both are strongly motivated by my combined background in computer science & engineering and music:

• The algorithmic surfacing of information that users would not discover by themselves. Our present-day search engines and recommender systems strongly focus on replicating earlier evidenced consumption success. But what if a user would want to develop a new interest? And what about those many items that got digitized, but hardly ever get found, simply because too few people know of their existence? As a musician, I frequently have been experimenting with this, and I believe the solution lies in proper presentation, contextualization, and comparative differentiation with respect to known standards. As a computer scientist, I am working on scaling and generalizing these thoughts, in the music domain and beyond.

• Validation and validity in data science: are we measuring and predicting what we intend to? In the current era of big data, we can acquire and analyze more data than ever, but this data is unstructured and messy, and measurement procedures may not have been optimal. Even more strongly, in many human-focused use cases, we may not be able to fully articulate what and where to measure, even though we have a good sense on what is an intended or unintended outcome. In music, we frequently encounter such challenges of measurement. Music information can digitally be described in many ways using many modalities, but the success of a song is typically determined by implicit human responses. As a computer scientist, I am interested in developing validation techniques that give us more confidence in our measurement procedures, also when they occur ‘in the wild’, outside of fully controlled lab settings. In this, I also explicitly am inspired by notions of psychometric validity in the social sciences domain.