Home

The purpose of this application is to index some documents with the AGROVOC Thesaurus (the default is English). Starting from a list of URLs of TXT, HTML or PDF documents, the application generates an [RDF file] (https://github.com/agrisfao/agrotagger/wiki/Example-of-AgroTagger-output).

The application is composed of three Java sub-applications that need to be executed sequentially. The output of the application is a tar.gz file, stored in the output location defined by the user as "output directory without the ending slash".

Information about Java sub-applications and examples of usage are available in the Java Applications page.

Information about bash scripts to execute JAVA applications is available in the project description.

The AgroTagger is entrirely based on AGROVOC. If you want to re-train the internal model (based on MAUI) or if you want to update AGROVOC or to use it in a different language, instructions are available in the New AGROVOC page.