Software

Meeting recorder and analysis tools

Helsinki Institute for Information Technology (HIIT) released Meeting Recorder on 24 August 2016, which is a tool for recording meeting video and audio. It supports multiple video feeds and writes directly to disk (preventing e.g. a power failure from destroying the entire recording). In parallell a set of analysis tools were released for tracking objects and gaze in meeting recordings.

semigeom

The Finnish Institute of Occupational Health (FIOH) has released the semigeom software package for the R statistical computing language on 28 June 2016. The semigeom package provides functions for performing semigeometric tiling of data matrices. The basic idea is to find regions in data matrices where the density of some feature, e.g., ones in a binary matrix, is large enough, as specified by some threshold. The semigeom R-package is licensed under the MIT-license. The details of the algorithm are described in the paper: Andreas Henelius, Isak Karlsson, Panagiotis Papapetrou, Antti Ukkonen and Kai Puolamäki. Semigeometric Tiling of Event Sequences. In proceedings of the European Conference on Machine Learning and Knowledge Discovery in Databases, 2016. (To appear.)

COCOREG

The Finnish Institute of Occupational Health (FIOH) has released the COCOREG software package for the R statistical computing language on 3 May 2016. The cocoreg algorithm extracts shared variation from a collection of datasets using regression models. The package is published under the MIT license and can be downloaded from the CRAN package repository. The details of the algorithm are described in the paper: Korpela J, Henelius A, Ahonen L, Klami A, Puolamäki K (2016) Using regression makes extraction of shared variation in multiple datasets easy. Data Mining And Knowledge Discovery, accepted for publication in 2016.

Digital Me server

Helsinki Institute for Information Technology (HIIT) released Digital Me (DiMe) server version 0.1 on 25 January 2016 together with the first of many DiMe loggers to come: Mac Desktop Logger version 1.0.2.  The newest version 0.2 was released 30 November 2016. The idea of DiMe is to collect your personal data from various loggers into a central place that you control. The current version of the server runs on Mac OS X, Linux and Windows. See the DiMe web page for more information.

Maui 2.0

Helsinki Institute for Information Technology (HIIT) released Maui 2.0 on 10 October 2014. Maui 2.0 is a keyword extraction tool, built on the old Maui, a topic indexing tool with several improved added. The source is released under GPLv3 and can be accessed from the HIIT github repository.

Spectral Mass

Helsinki Institute for Information Technology (HIIT) released Spectral Mass on 6 October 2014. Spectral Mass is an EEGLAB Extension used to compute the probability that two EEG datasets are similar in terms of both power (ERSP in EEGLAB terms) and coherence (ITC). The source code is released under the GPLv2 license and can be accessed from the HIIT github repository, for more information see the Spectral Mass web page.

MIDAS

The Finnish Institute of Occupational Health (FIOH) has released the MIDAS (Modular Integrated Distributed Analysis System) system on 13 August 2014. MIDAS is a system for online analysis of streaming physiologic signals and other time series and for easy integratration of features (e.g., psychophysiological metrics) into online machine learning frameworks. The MIDAS system consists of distributed processing nodes. A central node implements as RESTful JSON API which can be used to access information from the processing nodes. The MIDAS system is published under the MIT license and can be downloaded from github. More information can be found in the MIDAS wiki.