About the Tutorial

In recent years it has been realized that many data mining and machine learning problems, especially unsupervised ones, can be formalized as discrete constraint satisfaction and optimization problems. This tutorial provides a detailed overview of the use of constraint solving technology, such as constraint programming, SAT solvers, and MIP solvers, to solve these problems. After a short introduction to different types of solvers, we will provide an overview of a number of different machine learning and data mining problems and how they can be solved using these solvers. While we will discuss how standard data mining problems, such as itemset mining problems, can be solved using such solvers, the focus of this tutorial will be on the use of these solvers on data mining problems that go beyong itemset mining, such as clustering, block modeling, sequence mining, and decision tree induction, and we will provide an overview of recently proposed constraints on fairness and the interpretability of models.


The slides of the tutorial will be available here. For an impression, check our earlier IJCAI tutorial. However, the ECMLPKDD tutorial will be focused differently, providing more focus on topics such as finding explainable models, fair models and approaches based on MIP solving.


Siegfried Nijssen

Siegfried Nijssen is assistant professor at the Université catholique de Louvain. Together with Tias Guns, he developed the CP4IM framework for mining itemsets using constraint programming. He published numerous other papers concerning the discovery of patterns in data and the use of declarative languages in data mining. He has published at KDD, ICDM, ECMLPKDD, ICML, IJCAI and ECAI, among others. He was program chair of ECMLPKDD in 2013, is senior PC member of KDD and ICDM, associate editor of the KAIS journal and editor for the DMKD and ML journals.

Tias Guns

Tias Guns is an Assistant Professor at the Vrije Universiteit Brussel (VUB) and a fellow at the DTAI lab of the KU Leuven. His research lies on the border between data mining and constraint programming, and his main interest is in combining methods from both fields. As part of his PhD, he has developed the CP4IM framework which showed for the first time the potential of using constraint programming for pattern mining. His PhD was awarded with both the constraint programming dissertation award and the ECCAI artificial intelligence dissertation award. He is an active member of the community and has organized a number of workshops and a special issue on the topic of combining constraint programming with machine learning and data mining. His PhD was supervised by Nijssen and Luc De Raedt.

Ian Davidson

Ian Davidson is Professor at the University of California, Davis. His research interests span adding constraints to data mining algorithms (constrained clustering), human-in-the-loop learning (active and transfer learning) and more recently formulating data mining problems in constraint programming languages. He has published approximately a dozen papers in AAAI/IJCAI including two on the topic of this tutorial (2016, 2017). He has been an area chair for all the leading data-mining conferences for the last five years and is on the editorial board of ACM Transactions of Data Mining, IEEE Transactions of Knowledge Data and Engineering and Springer's Journal of Data Mining. He has known Guns and Nijssen for over five years and have presented at various summer schools and conferences on the topic of the tutorial with them.