Department of Computer Science
Mats Sjöberg

Mats Sjöberg

Postdoctoral Researcher, D.Sc. (Tech.)
CoSCo / INTENT research group
Helsinki Institute for Information Technology HIIT Mats Sjöberg

Contact information

Room A341, 3rd floor of the Exactum building,
Gustaf Hällströmin katu 2b,
Kumpula campus area,
Helsinki, Finland
Postal Address:
Department of Computer Science,
P.O. 68, FI-00014,
University of Helsinki,
Email: I'm moving to Aalto CS department 1.1.2018, please use my permanent email address: (GPG: 99E8E6D3)

Research interests

My research focuses on machine learning applications, especially programming computers to learn to recognise patterns in different kinds of data. I have previously worked with huge image and video datasets, in particular detecting affective and emotional content, and general concept detection. Currently my focus is on personal data, in particular creating machine learning tools for individuals to manage and understand their digital footprint.

Current topics

Earlier topics



Digital Me

The idea of the Digital Me (DiMe) server is to collect your personal data from various loggers into a central place that you control. Our focus is on tracking knowledge work and developing applications that can help the knowledge worker manage their working life better.

[Source code] - the source code is freely licensed (MIT).
[QE2017 slides PDF] - my slides from the Quantified Employee 2017 event explaining DiMe and time tracking.
[MyData 2016 slides PDF] - my slides from the MyData 2016 conference explaining Re:Know and Digital Me.
[Pre-print PDF] - Digital Me position paper presented at the Symbiotic 2016 workshop.

Selected publications

Full list of publications in the TUHAT database.

Google Scholar profile.

My ORCID is 0000-0002-3157-7668.

Doctoral thesis

From pixels to semantics: visual concept detection and its applications.
Department of Information and Computer Science, Aalto University School of Science.
[Available online]

Master's thesis

Content-based retrieval of hierarchical objects with PicSOM
Laboratory of Computer and Information Science, Helsinki University of Technology.
[Abstract PDF] [Entire document PDF].