Capability Case:
Automated Content Tagger
Intent:
To provide a way for people to add annotations to electronic content. Annotations can be comments, notes, explanations and semantic tags.  Content annotation provides meta-data that is becoming increasingly important to improve the precision of search as well as context-based information retrieval and repurposing.

Solution Story:
AeroSWARM Automated Markup, HighWire Stanford e-Library

AeroSWARM finds references in a news story, and renders them as semantic mark-up

The creation of markup from unstructured text sources such as web pages is tedious and time-consuming. Anyone who produces documents on a regular basis (e.g., intelligence analysts, commanders), or who has a large quantity of legacy documents, needs some form of automated markup assistance. Lockheed Martin has built a tool called AeroSWARM, which reduces the effort required for markup. It automatically generates OWL markup for a number of common domain-independent classes and properties. The author can then manually do markup additions and corrections to the output of AeroSWARM. 

A user can specify the set of web pages to mark up, and choose a target ontology.  Then, AeroSWARM generates OWL markup like that shown in the figure above. The sample markup includes entities (e.g., person, place, organization), relations (e.g., Pinochet persToLoc Santiago) and co-references (e.g., Pinochet sameIndividualAs Augusto Pinochet). A table on the AeroSWARM site describes all the entities and relations that can be automatically identified and marked-up.

Benefits:
Reducing the workload for annotated markup to a level manageable by human effort.  The markup enables valuable business capabilities such as improved search, information discovery and retrieval.

Featured Products:
AeroSWARM from Lockheed Martin

Applicable Products:
ClearTags from ClearForest, Enigma 3C Platform from Enigma, Semio Tagger from Semio, SKIP from SemanTx Life Sciences, Stratify Discovery System from Stratify