The EDA group at LIACS develops algorithms and theory that enable domain experts to describe data, discover causal relationships, 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, such as understanding the causal structures that govern the data generation process. The goal of the latter is to make intelligible, causal, and 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 (often using the minimum description length (MDL) principle), causal inference algorithms, statistical pattern mining, and interactive data exploration and analytics. 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.