csteed / eden

The Exploratory Data analysis ENvironment is an interactive data exploration tool for hypervariate data.

Exploratory Data analysis ENvironment (EDEN)

EDEN is an interactive visual analytics tool for exploring quantitative multivariate data. EDEN is written in Java and runs on Mac OS X, Windows, and Linux operating systems. EDEN is developed and maintained by the Oak Ridge National Laboratory Computational Data Analytics Group. The lead developer is Chad A. Steed.

If you are using EDEN for your work, we would greatly appreciate you citing the following paper:

Chad A. Steed, Daniel M. Ricciuto, Galen Shipman, Brian Smith, Peter E. Thornton, Dali Wang, and Dean N. Williams. Big Data Visual Analytics for Earth System Simulation Analysis. Computers & Geosciences, 61:71–82, 2013. doi:10.1016/j.cageo.2013.07.025 http://dx.doi.org/10.1016/j.cageo.2013.07.025

Compiling the EDEN Source Code

Compiling EDEN is straightforward. The first step is to clone the repository. We supply a Maven POM file to deal with the dependencies. In the Eclipse development environment, import the code as a Maven project and Eclipse will build the class files.

EDEN requires the DataTable library. The DataTable library must first be compiled and packaged into a jar file. Instructions for compiling DataTable are available on the DataTable repository. After compiling DataTable into a jar file, the jar file must be installed in Maven. Using the following command will install the DataTable file in the local Maven repository:

$ mvn install:install-file -Dfile=datatable-1.0.0.jar -DgroupId=gov.ornl.datatable -DartifactId=datatable -Dversion=1.0.0 -Dpackaging=jar -DgeneratePom=true

Optionally, the DataTable jar file can be specified with a file path in the pom.xml file instead of installing it in the local Maven repository.

To compile EDEN on the command line, issue the following commands:

$ mvn compile
$ mvn package

Running EDEN

These commands will generate 2 jar files in the target directory. Copy the jar file with dependencies into the scripts directory and run either the eden.bat script (Windows) or the eden.sh script (Mac or Linux). The EDEN GUI should appear after issuing this command. Example data files are provided in the data directory for trying EDEN out.

Precompiled EDEN jar Files

Earlier precompiled releases of EDEN are available from this site under the releases section.

eden's Issues

Zoomable focus lens method

Add comparative analysis features (visually build multiple selections of lines for comparison)

Multiple range queries on each axis

Add capability to create more than one range query for each axis.

Animate aerial perspective shading on an axis

Add ability to select an axis and color scale for the polylines.

Add ability to select some polylines and spawn a new parallel coordinates panel (new window)

Add ability to select individual polylines

Aerial Perspective Shading

NullPointerException when removing columns

If "Show Statistics for Selected Lines" is checked and a range is selected on at least one of the axes, a null pointer exception is thrown and functionality breaks when you remove a column (disable it).

Tested on version: 0.11.2

Ability to normalize axes to share the same min/max values

Either a menu item to select some set of axes to find the common min/max values and set the min/max for all axes to be the overall range. This will help with comparisons of axes that are of the same units.

Add ability to show polyline details in a table view

Show linked 3D or 2D view of the data (when x, y, z variables are present)