Introduction to METS

What is METS?

The Metadata Extraction and Tagging Service (METS) is a Department of Defense Intelligence Information System (DoDIIS) Core Service that extracts information found within unstructured documents. This promotes integration with structured data and will significantly improve search, analysis, and knowledge discovery. METS automates the normalization of, and extraction of information from, text documents, making the content of the documents quickly available as XML and OWL (Web Ontology Language) / RDF (Resource Description Framework) for intelligence analysis.


METS has been developed in support of the client's Strategic Plan (Fiscal Years 2004-2009), the "Content Tagging Way Ahead" White Paper and Proposal, Presidential Budget Decision-339C2 for Content Tagging, IC Program Guidance -- FY2006-2011, and Presidential Information Sharing Directive EO 13355.

How do I Use METS?

METS provides a SOAP web service. You can submit documents to METS by running client software which is written to communicate with the METS web service via SOAP. Or you can use a RESTful web service. You can also use a METS web page which communicates indirectly to that service as well, to submit documents directly and view the results in a browser.

For security reasons, the METS web service uses one-way SSL.

Details on how to implement a METS web client or use the web page can be found in the METS ICD and Users' Guide.

What is METS' Background?

The work on what has become METS was started mid-2003 as an project to evaluate:

  • the benefits of "tagging" message traffic using XML
  • the coverage and accuracy of various commercial extraction tools

Upon conclusion of that project, the development team was asked to turn the evaluation system into an operational one to serve the community.

What is METS' Status?

METS has been available as an operational capability since the 1.0 deployment was approved and fielded in February 2005. Release 5.0 was fielded in late February 2012; followed by the 5.0.1 patch release in early July 2012. Release 5.5 (of which this version of the website is a part) was deployed Sep 1 2012. Development of release 5.6 is underway, with deployment planned for late February 2013. For more details, see here

What are METS' Features?

METS accepts text documents in the usual formats (including Word and PDF) and encodings, normalizes each into a standard XML form using UTF-8, identifies the document metadata, and extracts entities, events, and relations. The results are represented in a range of user-selectable XML forms: PUBS v5, v7, or v9, DDMS v3 or v4, NDMS v2, and OWL/RDF/XML.

METS provides SOAP and REST web services which accept a document, process it, and return the results along with any Warnings and Error. It also provides a web page which supports browsing for and submitting documents and viewing the results.

For the details see here.

How do I try METS Out?

The unclassified METS 5.5 test web interface and test web service are available for testing at:
The web interface
The SOAP web service
The REST web service

The contents of the METS website are ©2008-2012 by Booz Allen Hamilton and written by Richard Lee.