I research and build human-centric machine learning systems. At OTTO, I focus on forecasting that combines algorithmic predictions with human judgment. My PhD from TU Delft explored viewpoint biases in web search results; specifically how to measure them and their effects on user behavior (dissertation ↗). With a background blending computer science, psychology, and research methods, I'm drawn to hybrid intelligence, the space where technical systems enhance human decision-making.
CVA comprehensive flowchart to help practitioners to find the right statistical test for their data and hypotheses. Used in university classes all over the world.
ECIR 2024
We argue that web search can and should empower users to form opinions responsibly. Building on digital humanism, a perspective focused on shaping technology to align with human values and needs, we identify challenges and research opportunities that focus on the searcher, search engine, and their complex interplay.
HCOMP 2023 · Best Paper Honorable Mention
We propose a Responsible AI (RAI) methodology designed to guide the data collection with a set of metrics for an iterative in-depth analysis of the factors influencing the quality and reliability of generated data. We propose a granular set of measurements to inform on the internal reliability of a dataset and its external stability over time.
CHI 2023 · Best Paper Award
A user study investigating the individual and combined effects of explanations, human oversight, and contestability on informational and procedural fairness perceptions for high- and low-stakes decisions in a loan approval scenario.
CHIIR 2022 · Best Paper Award
We propose a novel, two-dimensional way of representing viewpoints that incorporates a viewpoint’s stance degree as well as its logic of evaluation, allowing for more meaningful analyses and diversification interventions compared to current approaches.
HCOMP 2021 · Best Paper Award
We demonstrate that potential effects of cognitive biases are often ignored in crowdsourcing studies. To amend that, we propose a 12-item checklist adapted from business psychology to combat cognitive biases in crowdsourcing.
SIGIR 2021
A user study investigating whether users adopt the viewpoints promoted by higher-ranked documents in web search on debated topics.
Hypertext 2021 · Best Paper Award
A user study investigating whether search result obfuscation can mitigate confirmation bias in web search on debated topics.