I am a computer scientist who likes algorithm design, machine learning and programming.
After spending 15+ years in academia, I thought it was time to do something else for a change. As of Jan 1st 2020 I work at Speechly. Occasionally I might also teach and do a bit of research. I am also a Docent (of data science) at the Department of Computer Science, University of Helsinki.
Here you can find reference implementations of some algorithms I've worked on, and datasets used in some of my papers.
Triplet data that we collected for the experiments in (Heikinheimo & Ukkonen, HCOMP 2013) is available here. Please see the README file for instructions. If you use this in your research, I am kindly asking you to cite our HCOMP 2013 paper.
Spectra is an algorithm for quickly estimating the "pattern frequency spectrum" of a binary dataset. This is a curve that shows the number of frequent itemsets for a given support threshold.
Please try out a demo, it will work even on your (smart)phone! Source code is available on GitHub.
The paper (van Leeuwen & Ukkonen, ECMLPKDD 2014) is available online, and more details about the demo in particular can be found in our short technical report.
Fast-Skyline is an algorithm for computing approximate “skylines” (non-dominated sets) of subsets of size-k subject to two functions, one linear, one submodular. That is, the algorithm computes the set of non-dominated subsets of size-k. This problem has applications in e.g. viral marketing.
The algorithm is described in (van Leeuwen & Ukkonen, SDM 2015), and extends the work in (van Leeuwen & Ukkonen, ECMLPKDD 2013). More information can be found here.
In our ECMLPKDD 2017 paper we describe a method for ranking subgroups using network-based techniques. The code is available on GitHub.
Antti Ukkonen, Chief Product Officer
antti (at) speechly.com (for work related matters)
antti.ukkonen (at) helsinki.fi (for teaching and research related things)
antti.ukkonen (at) gmail.com (for other business)
My LinkedIn page