Current version of CIVET with document control, annotation/extraction and data entry templates
Switch branches/tags
Nothing to show
Clone or download
Pull request Compare This branch is 2 commits behind civet-software:master.
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Failed to load latest commit information.
djcivet_site
.gitignore
CIVET.Documentation.pdf
LICENSE
README.md

README.md

CIVET-Django Beta 0.9

CIVET -- Contentious Incident Variable Entry Template -- is a customizable data annotation, extraction and entry system running in the Django web framework. CIVET is being developed by the NSF-sponsored project titled "A Method for Leveraging Public Information Sources for Social Science Research" which is creating tools to improve the efficiency of data generation in the social sciences. The project has an initial focus on coding event data in the domain of contentious politics, but we expect that these tools will be relevant in a number of data-generation domains.

CIVET currently has the following features

  • A YAML-based document management format designed for storing collections of relatively short texts such as news articles together with metadata, a coding form, and cases coded from these texts. We expect to build a series of additional tools using this format.
  • A system for creating web-based data coding forms with standard HTML components such as checkboxes, pull-down selection menus and text entry, without the need to know HTML or web programming
  • Automatical annotation of texts to identify named entities, numbers and user-specified categories.
  • Integration of the annotated texts with the web form so that data fields can be extracted by going through a specific set of annotated words. Simple data entry and select/paste from the texts are also supported.

The file CIVET.Documentation.pdf provides information on installing and operating the system. In the current version, Beta 0.9, all of the major functions needed for specifying a form, readings collections of texts, annotating and coding the text, and writing the resulting data are operational.

While the system has been tested during the software development process, it has yet to be tested in the field, so you should experiment to make sure it is coding and saving your data correctly, and then check the output files periodically to make sure it is still working. We expect to start field testing in September-2015.

###Acknowledgements The development of CIVET is funded by the U.S. National Science Foundation Office of Multidisciplinary Activities in the Directorate for Social, Behavioral & Economic Sciences, Award 1338470 and the Odum Institute at the University of North Carolina at Chapel Hill with additional assistance from Parus Analytics.

For bug reports, suggestions and further information, contact schrodt735@gmail.com