Automatic Window Width Determination for CoCoNAD

Download

Description

The document width.pdf contains some result diagrams for experiments in which it was tried to find the analysis window width to be used with CoCoNAD automatically from the given data set. The ideas underlying the approach are described in the paper [Picado-Muiño and Borgelt 2015].

The archives width.{zip,tar.gz} contain scripts and other source files, with which the experiments were conducted and the document with the result diagrams was created.

Note that the scripts etc. were developed on/for a GNU/Linux system (Ubuntu 14.04) and thus are directly executable on such a system or a similar one (that is, some other GNU/Linux distribution). Although at least most of the Python scripts should also be working on a Windows system (with the possible exception of the parallelization scripts), most of the other scripts (like the run script, which is the main control script, and the makefile, which controls generating the diagrams from the result data) may need porting to batch files or something similar.

On a GNU/Linux system, the following software needs to be installed to run the experiments:

• Python, preferably version 2.7.x (although version 3.x can be made to work as well),
• the PyCoCo extension module for Python, which makes a fast C implementation of the CoCoNAD algorithm available as a Python function (the scripts in the cocoexp package will also work without this extension module, namely by falling back on a pure Python replacement, which, however, is slower by a factor of about 40 or more),
• a LaTeX system, preferably TeX Live (which is the standard LaTeX system for Ubuntu and many other Debian based GNU/Linux systems and may be installed directly through the package manager), in particular the pdflatex program,
• the MetaPost program for the LaTeX system (which is usually already contained in standard LaTeX installations, but on some systems may require separate installation), in particular the mptopdf command,
• some standard Unix tools, which are usually available on all Unix systems even with a very basic installation (like bash, awk, tar etc.) and are easy to install otherwise.

On such a system the experiments can be run by simply calling the main script run (in the directory width) on the command line, which does everything. The execution of the experiments exploits 4-fold parallelization, thus making full use of the quadcore processors basically all modern computers are equipped with. The progress of the experiments can be followed on the command line, to which regular progress messages are written. Once all experiments are completed (which should take less than an hour on a modern computer system), the result diagrams are created and compiled into the final documents, which are also directly available above.

References

• Automatic Learning of Synchrony in Neuronal Electrode Recordings
  David Picado-Muiño and Christian Borgelt
  Proc. 16th Congress Int. Fuzzy Systems Association (IFSA)
  and 9th Conf. Europ. Society for Fuzzy Logic and Technology (EUSFLAT)
  (IFSA-EUSFLAT 2015, Gijón, Spain), accepted, to appear.
• Finding Frequent Patterns in Parallel Point Processes
  Christian Borgelt and David Picado-Muiño
  Proc. 12th Int. Symposium on Intelligent Data Analysis, 116-126.
  Springer-Verlag, Berlin/Heidelberg, Germany 2013
  ida_13.pdf (443 kb) ida_13.ps.gz (491 kb) (11 pages)
• Frequent Item Set Mining for Sequential Data: Synchrony in Neuronal Spike Trains
  David Picado-Muiño and Christian Borgelt
  Intelligent Data Analysis 18(6):997-1012.
  IOS Press, Amsterdam, Netherlands 2014
  doi:10.3233/IDA-140681 iospress.com
  (16 pages)
• Finding Neural Assemblies with Frequent Item Set Mining
  David Picado-Muiño, Christian Borgelt, Denise Berger, George Gerstein, and Sonja Grün
  Frontiers in Neuroinformatics 7:article 9
  Frontiers Media, Lausanne, Switzerland 2013
  doi:10.3389/fninf.2013.00009 frontiersin.org
  accfim.pdf (1797 kb) accfim.ps.gz (772 kb) (14 pages)
• Statistical Evaluation of Synchronous Spike Patterns extracted by Frequent Item Set Mining
  Emiliano Torre, David Picado-Muiño, Michael Denker, Christian Borgelt, and Sonja Grün
  Frontiers in Computational Neuroscience, 7:article 132
  Frontiers Media, Lausanne, Switzerland 2013
  doi:10.3389/fncom.2013.00132 frontiersin.org
  (13 pages)