AgroTagger is open source.

It is not under active development by FAO at present, but the source code is available on Github: https://github.com/fcproj/agrotagger. It is recommended to rely on a programmer or someone who knows how to run JAVA applications from the command line.

The AgroTagger was developed in the context of a couple of EU projects, such as agINFRA and SemaGrow. TheAgroTagger was used to enrich data in different digital information environments (AGRIS is among them) and to index documents with AGROVOC multilingual thesaurus. 

NOTE:

-- AgroTagger is not under active development by the FAO of the UN at present !

• Anyone in the world can take the source code from Github: https://github.com/fcproj/agrotagger and add new features;
• You cannot test the software on the web. It's an application, not a website; 
• Technically, it's a JAVA application, so it should run also on Windows : the application on Github is provided with scripts for Linux, but anyone can write their own scripts for Windows or MacOS;
• The AgroTagger on GitHub is a Java Command Line application which supports only English. The application is language independent, but the model created is based on AGROVOC in English. If one is able to build a new MAUI model, it can support all the languages.

More information about AgroTagger in this blog post:

http://aims.fao.org/activity/blog/agrotagger-classifier-leveraging-knowledge-encoded-agrovoc-agris-items