GeoCrunch is a tool to extract the spatial structure of digital data collections.
Switch branches/tags
Nothing to show
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Failed to load latest commit information.
.externalToolBuilders
.settings
config
ner_models
src
webapp
.classpath
.gitignore
.project
LICENSE.md
NOTICE.md
README.md
bachelor_thesis.pdf
geoCrunch.xml
pom.xml
starter.bat
starter.sh
viewer.png

README.md

GeoCrunch Quick Start Guide

What does it do?

GeoCrunch extracts, collects and visualizes spatial data and metadata from file collections of any media type. Though metadata can be retrieved from all file types, retrieval of information from file content bodies is limited to .csv and natural language .pdf files, yet. Supported languages include English and German.

The software consists of

  1. an extraction tool and
  2. a web browser application for visualization of the results.

Extraction is performed on three levels:

  1. file or folder names are evaluated for place names
  2. metadata is examined for coordinates or place names
  3. analysis of file content bodies on .csv (coordinates) or .pdf (place names)

GeoCrunch screenshot

Getting started

See the analysis results of sample data in the GeoCrunch viewer

You may want to visit the GeoCrunch Viewer life demo.

Alternatively, download and unpack the demo.zip package. Start the GeoCrunch Viewer by opening the file demo/webapp/index.html in a standard conform web browser (I recommend Firefox. Safari was reported to have problems, Chrome cannot handle local file resources correctly). An internet connection is required to render the map tiles.

Browse the collection tree by clicking on the items on the left, see how the map display changes. Click on any marker on the map. The related file entry becomes highligted in the directory tree and a popup on the marker shows additional information on the context.

Click on the button 'Filters' on top of the collection tree. Select the filter options you would like to see.

Create a spatial analysis of your own data collection

Remove the contents of the folder demo/webapp/data and insert the data you wish to analyse. There are restrictions on the amount of geocoding requests towards the various services, thus it is recommended to analyze no more than ~10 scientific .pdf papers at once.

Have a look at the configuration file user.properties at GeoCrunch/config and adjust according to your needs. ATTENTION: geocoding webservices have been turned off in that configuration file, as re-processing the provided example data would block the geocoding query limits of the provided API keys uneccessarily for at least one hour (for you and others). Turn this on only, if in fact you want to process new data. Consider creating your own API keys for the utilized geocoding services.

The file default.properties holds the default settings suitable for most tasks and should not be changed in order to easily revert all adjustments by replacing user.properties by a fresh copy of default.properties.

Open up a terminal and type

java -jar geoCrunch.jar

Alternativey, just click on one of the starters suitable to your operating system. Check the results in the GeoCrunch viewer, after the process has terminated.

More information

Please read the related bachelor_thesis.pdf to learn about the scientific backgrounds and concerns of the software development.

The full application sourcecode is available here: https://github.com/hashmich/GeoCrunch.git

For further questions, send an email to the developer: mail@hendrikschmeer.de