This week I participated the 34th IEEE International Conference on Data Engineering (ICDE 2018) in Paris, France, together with my colleagues and co-authors Jefrey Lijffijt and Wouter Duivesteijn. I had two papers in poster session, the first presented by me and the second by Jefrey:

  1. Kai Puolamäki, Emilia Oikarinen, Bo Kang, Jefrey Lijffijt, Tijl De Bie. Interactive Visual Data Exploration with Subjective Feedback: An Information-Theoretic Approach. In Proc ICDE, 1208-1211, 2018. Poster, extended version at arXiv:1710.08167 [stat.ML]

  2. Jefrey Lijffijt, Bo Kang, Wouter Duivesteijn, Kai Puolamäki, Emilia Oikarinen, Tijl De Bie. Subjectively Interesting Subgroup Discovery on Real-valued Targets. In Proc ICDE, 1352-1355, 2018. Extended version at arXiv:1710.04521 [stat.ML]

Presenting poster at ICDE 2018 At Eiffel tower

ICDE is the premier conference in the field of data engineering. It was quite an interesting experience for me, because my own computer science background is in the field of computational data analysis - the relevant keywords being machine learning, data mining, data science, and artificial intelligence - but not on the data engineering nor databases.

One of the pains in being in growing and continually developing field is fragmentation of research topics and people to different sub-disciplines. This was visible also in the ICDE: in addition to the traditional core topics of the conference there were many papers that could as well have been presented, e.g., in data mining or visualisation venue. This applies to our papers above as well, topics of which are relevant also for, e.g., the data mining and visualisation communities.

Generally speaking, it happens sometimes that the same problems are being worked on neighbouring disciplines - often from different starting points - without people knowing and benefiting from each others’ work. It is clear that all parties would benefit from better inter-disciplinary communication. There is nothing new here, we tried to solve the problem with respect to visual analytics already almost a decade ago; still waiting for the solution…

As an added benefit, the spring in Paris is substantially warmer than the spring in southern Finland. I spent two nights in Paris and in my mornings there I ran a total of 23 kilometers, on both days in good time to make it to the keynote talks that began at 8:30. :) Before 7 o’clock in the morning the streets next to the Seine river were quite occupied by the runners, after which the sun rose and the rest of the Paris seemed to really wake and warm up.