Polina Rozenshtein

Polina Rozenshtein · Tue, 05.12.2017


This September I attended ECML-PKDD 2017 The European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases, which took place in Skopje, Macedonia. I have presented our accepted paper titled The network-untangling problem: From interactions to activity timelines. This paper is also a part of my PhD studies.

ECML-PKDD is the largest joint European conference on Machine Learning and Data Mining, which brings together ML and DM communities. While being narrow in topics, parallel sessions contained both ML and DM talks, which ensured exchange of the latest ideas on the active research topics of these adjacent fields. There were several sessions on various aspects of Graph Mining, which focused on the recent work directly related to my research topic. There were also several brilliant tutorials and invited talks. I particularly enjoyed tutorial on Machine learning with fossil data by Indrė Žliobaitė from Helsinki University and Earth observation talk by Pierre-Philippe Mathieu from ESA.

The conference overall was highly related to my studies and I would like to thank the HICT travel stipend for supporting my attendance.