Description
The compositions and the robCompositions packages are software extension of the free and open-source statistical environment R. They include multivariate methods for the analysis of compositional data. However, subject matter specialists in, for example, geosciences or geochemistry often do not have any knowledge in programming or R. Thus, the aim is to provide the main facilities of both packages within a point and click interactive environment. For this task, the package Shiny should be used to turn analyses into interactive web applications that anyone can use. Several steps have to be considered:
- interaction: call the workhorse computational functions of robCompositions and compostions from shiny. Either write smart wrappers for it or revise the object-oriented class structure of these packages.
- extend these functionalities by some basic multivariate statistical methods and plots. As a guideline, use the book of Reimann, Filzmoser, Garrett and Dutter (2008).
- interactive features: shiny is favorable for its possiblity to create interactive applications. For every method, some interactive elements (e.g. sliders or text input) should be included.
Benefits for the Student
The student will get deep knowledge in programming in R and interactive web applications with shiny but also basic knowledge in compositional data analysis.
The student will work together with experts in this field of research and the project outcome (if successful) will be made available, distributed over the web and spread to the community.
Requirements
Basic knowledge in multivariate statistics. Advanced knowledge in any scripting language and especially in R. Knowledge in programming web applications.
Mentors
Matthias Templ, Peter Filzmoser
Contact
Mentors are regularly around in our GSoC IRC channel #TU-CSE-SoC at irc.freenode.net. You can also reach us via the mailinglist – send an email to This email address is being protected from spambots. You need JavaScript enabled to view it. using the prefix [VIM] (a subscription is required).
Links
www.r-project.org
http://cran.r-project.org/web/packages/robCompositions/index.html
http://cran.r-project.org/web/packages/compositions/index.html
http://www.rstudio.com/shiny/
More information
http://www.iue.tuwien.ac.at/cse/wiki2014/doku.php?id=shiny_on_compositional_data_analysis