Open research data and software
Statistical ecology and machine learning
DMT Dependency Modeling Toolkit. Probabilistic tools for dependency analysis between multiple data sources (R/CRAN). Probabilistic PCA, factor analysis, CCA, regularized variants, dependency-based dimensionality reduction etc. ICML/MLOSS workshop, Israel 2010.
netresponse Modeling context-specific activation patterns in genome-wide interaction networks (R/Matlab). Originally applied to study transcriptional responses in genome-scale interaction networks across organism-wide collections of gene expression data. doi:10.18129/B9.BIOC.NETRESPONSE
Digital Humanities and Computational Social Science
rOpenGov R package ecosystem for open government data analytics (R/GitHub). introduced in NIPS Machine Learning Open Source Software workshop 2013.
COMHIS Helsinki Computational History Group data analytics infrastructure.
Various data sets related to publications available in the following open repositories:
Full agreement texts for academic publisher agreements. The agreements were released by the FinELib consortium of research libraries following our FOI request in April 2018.
Scientific journal subscription costs in Finland 2010-2017; MoE/ATT. The data set was released by Finnish Ministry of Education (Open Science and Research Initiative) following my Freedom of Information Request (summary). Finland became the first country to systematically collect and release this information.