ProcessNetwork v1.4, January 16th 2015
Attribution and Licensing
The ProcessNetwork software was written between 2005 and 2008 at the University of Illinois at Urbana-Champaign, in the Department of Civil Engineering, funded 2006-2008 by a NASA fellowship #NNX06AF71H , and then continued development has occurred at Arizona State University, funded 2013-2015 by a grant from the NSF’s Macrosystems Biology program BIO-1241960. Dr. Benjamin L. Ruddell is the primary author of the software, but many collaborators have contributed to its ongoing development. The work is that of the author, and its accuracy and implications are not necessarily supported by the funding and employing organizations. The software copyright is held by Dr. Ruddell, but is shared for fair use with acknowledgement and citation of the author’s contribution under a Creative Commons Public License. Those using the software are requested to cite the software using the following three citations, and to communicate with the author regarding modifications, extensions, and applications of the software.
Citation for the software:
Ruddell, B.L. (2008), ProcessNetwork Software, version (INSERT VERSION HERE), accessed at (INSERT URL OR SOURCE HERE) on (INSERT DATE HERE).
Citations for the methods employed in the software:
Ruddell, B. L.*, and P. Kumar (2009a), Ecohydrologic process networks: 1. Identification, Water Resour. Res., 45, W03419, doi:10.1029/2008WR007279.
Ruddell, B. L.*, and P. Kumar (2009b), Ecohydrologic process networks: 2. Analysis and characterization, Water Resour. Res., 45, W03420, doi:10.1029/2008WR007280.
Ruddell, B.L.*, N. Oberg, P. Kumar, and M. Garcia (2010), Using Information-Theoretic Statistics in MATLAB to Understand How Ecosystems Affect Regional Climates, MATLAB Digest Academic Edition, February 2010, www.mathworks.com/academia.
Dr. Praveen Kumar, for advising and co-authorship of publications at the University of Illinois
Dr. Ricky Robertson, for important histogram algorithm advice at the University of Illinois
Numerous students and colleagues who provided help with MATLAB coding
Cove Sturdevant and Dennis Baldocchi of the University of California, Berkeley for bug identification
Minseok Kang and Rong Yu for contributions to code versions 1.2 through 1.4
The purpose of this software is to delineate Dynamical Process Networks based on observations of information flow between discrete vector variables using information theory, including the Transfer Entropy statistic. The software also computes Shannon Entropy, Mutual Information, Relative Entropy, and many other information theoretic statistics. The software is valid for all kinds of data, but the typical application is to Dynamical information flow and Shannon Entropy, and the typical dataset a multivariate timeseries. The software is written for the MATLAB® scientific computing environment. This basic version of the software does not contain any visualization, plotting, or GUI features, but some of these features may be available separately. The basic version of the software contains a small set of preprocessing functions and filters, but others may also be useful.