The EDA group at LIACS, headed by dr. Matthijs van Leeuwen, develops algorithms and theory that enable domain experts to describe data and predict using interpretable patterns and models.

We focus on exploratory data mining and interpretable machine learning. The goal of the former is knowledge discovery, i.e., to gain new insights from data, whereas the goal of the latter is to make intelligible, data-driven predictions. Regardless of the objective, we aim for algorithms that are accurate, principled and reliable, and can be guided by the analyst (in order to exploit prior knowledge).

Our research builds on the state of the art in information theoretic data mining (e.g., using the minimum description length (MDL) and maximum entropy principles), statistical pattern mining, and interactive data exploration and analytics. More broadly speaking, our research can be situated in the fields of data mining, machine learning, data science, and artificial intelligence (AI).

We often work on interdisciplinary projects, as we strongly believe that this leads to interesting, fundamental data mining and machine learning problems for which the solutions we propose have the potential to positively impact society.