The Explorative Data Analysis research group at the Department of Computer Science of the University of Helsinki is lead by Prof. Kai Puolamäki. Our research interests include methods of data analysis and artificial intelligence to understand heterogeneous data sources and models, and to make analytics data-based decisions.

We are affiliated with the Finnish Center for Artificial Intelligence (FCAI), Helsinki Institute for Information Technology (HIIT), and Institute for Atmospheric and Earth System Research (INAR). We are the AI group in the Virtual laboratory for molecular level atmospheric transformations (VILMA) in the Academy of Finland Centre of Excellence Programme 2022-2029.

Ongoing research projects

Virtual laboratory for molecular level atmospheric transformations

We are the AI group in the Virtual laboratory for molecular level atmospheric transformations (VILMA): We will build digital twins that models aerosol formation processes and measurement devices and is powered by measured data, physical and chemical simulations, and probabilistic AI models. We will develop AI tools composed of explainable AI and visualization methods that can be used to study, understand, and build the digital twins. Furthermore, we will devise ways to quantify uncertainties inherent in data, models, and predictions.

The project is one of the new Centres of Excellence in Academy of Finland Centre of Excellence Programme 2022-2029.

Finnish Center for Artificial Intelligence

We are members of the Finnish Center for Artificial Intelligence (FCAI): A concrete step towards AI systems that are data-efficient, trustworthy and understandable. We jointly focus on creating ‘Real AI’ tools for AI-assisted decision-making, design and modeling. We develop AI techniques needed for systems that can help their users make better decisions and design better solutions across a range of tasks from personalized medicine to materials design.

FCAI is a joint venture by University of Helsinki, Aalto University, and VTT Technical Research Centre of Finland. It is funded in part by the Academy of Finland as one of the Finnish flagships for the period of 2019-2026.

Past research projects

Interactive Artificial Intelligence for Driving R&D

Our goal is a new generation of AI methodology that can drive a disruptive wave in the Finnish technology industry. The project develops AI software for white-box AI that better understands its users, making AI accessible to in-house domain experts, and improving efficiency and success rate.

The project is funded by the Technology Industries of Finland Centennial Foundation for the period of 2019-2021 (announcement in Finnish). The project is a collaboration with seven professors from Aalto University and University of Helsinki (including our group leader, Kai Puolamäki), the principal investigator being Prof. Samuel Kaski.

Human-guided data analysis

The current methods and processes of data analysis give the knowledge workers, who are rarely experts in data analysis, only a limited means to explore large heterogeneous data sets. We further develop and study the recently introduced formulation of the explorative data analysis task in terms of statistical significance testing and constraints to null hypothesis to develop novel methods of data analysis that are optimised for the use with humans and that can be controlled by the humans. The project has two use cases that are used to demonstrate the methods, namely analysis of scientific data sets collected at the Finnish Institute of Occupational Health and a prototype system by which medical doctors to analyse and study patient data.

Human-guided data analysis is an Academy Project funded by the Academy of Finland for the period of 2015-2019.

Structure from randomization

The objective of the project is to develop and apply statistically sound randomization methods to be used to find complex patterns from the data and that can be used in conjunction with the state-of-the art machine learning and data mining methods. By randomization we mean here a process by which we can create a controlled perturbation of the data. These perturbations can be used in statistical significance testing and to make the machine learning algorithms transparent and to explore the model space of machine learning algorithms. In this project we apply on data sets that are of relevance for work life.

Structure from randomization is an Academy Project in the ICT 2023 programme, funded by the Academy of Finland, during 2018-2019.