Greek Reuse Toolkit

The Greek Reuse Toolkit (GRT) is a toolkit designed for the retrieval of references in Ancient Greek (Koiné). It was developed for the ANR project Biblindex.

The toolkit performs four major functions:

• Reading ancient greek digital documents encoded in Unicode, BetaCode or BibleWorks code, and convert them into any of these formats,
• Managing a list of lemmatised forms,
• Retrieving references of a text into another in a customisable fashion (many parameters are available),
• Evaluating paradigms and methods for quotation retrieval.

The Greek Reuse Toolkit is currently only available in french.

Installing the Toolkit is as straightforward as unzipping the install package available below in any location (and making sure your java is up to date). Launch the application using GreekReuseToolkit.jar.

Quotation retrieval uses lemma lists that must be named and located in a very precise fashion. The install package already contains them.

Once launched, the application will ask you which task you wish to perform. Choose a task to be brought to the corresponding part of the software.

• To retrieve quotations, specify two texts and choose your parameters.

• The results are provided with a color/underline/bold scheme to quickly find out why a quotation was tagger as such.

• To evaluate quotation retrieval parameters, you have to specify two texts and a control set of quotations, then you can define as many experiments as you want before launching the experiment.

• The results are provided with precision and recall, both with a refined version that accounts for the self-quoting nature of biblical texts.

The Greek Reuse Toolkit is available under the GNU GPL v3.

Contributors: Samuel Gesche
Release date: 2014-06-30

• Software
• Netbeans Project
• Sample documents: our experiment sets (the documents in xml and their control quotation sets)