<!DOCTYPE article PUBLIC "-//NLM//DTD JATS (Z39.96) Journal Archiving and Interchange DTD v1.0 20120330//EN" "JATS-archivearticle1.dtd">
<article xmlns:xlink="http://www.w3.org/1999/xlink">
  <front>
    <journal-meta />
    <article-meta>
      <title-group>
        <article-title>Schema Management in a Semantic Wiki</article-title>
      </title-group>
      <contrib-group>
        <contrib contrib-type="author">
          <string-name>Semandex Networks</string-name>
        </contrib>
        <contrib contrib-type="author">
          <string-name>Inc. Princeton</string-name>
        </contrib>
        <contrib contrib-type="author">
          <string-name>New Jersey</string-name>
        </contrib>
        <contrib contrib-type="author">
          <string-name>bob}@semandex.net</string-name>
        </contrib>
      </contrib-group>
      <fpage>2</fpage>
      <lpage>3</lpage>
      <abstract>
        <p>A semantic wiki for tactical intelligence requires collaborative, iterative schema management. Application developers can provide users with control and visibility required to tailor an ontology to meet unique requirements.</p>
      </abstract>
    </article-meta>
  </front>
  <body>
    <sec id="sec-1">
      <title>Introduction</title>
      <p>
        Military and law enforcement operations rely on good tactical intelligence. That
intelligence is derived from information sources that vary depending on the location
and nature of the operation. Some data sources are highly structured and amenable to
computer processing; others are characterized by semantic and syntactic
incompatibilities that inhibit automated integration. These differences result in
underexploited data [
        <xref ref-type="bibr" rid="ref1">1</xref>
        ]. Emerging technologies can help intelligence analysts exploit
disparate datasets and better understand the tactical situation.
      </p>
      <p>One application that helps analysts organize and manipulate information from
disparate sources is a semantic wiki. A semantic wiki promotes rapid dissemination of
information and collaborative analysis in a distributed, virtual environment. It does so
by parsing information into a repository using an ontology, and by using that ontology
to help discover and represent additional relationships in the corpus of knowledge.
Different military and security situations deal with different types of information.
Changes in information type require changes in schema structure. One ontology does
not fit all situations. Accordingly, a semantic wiki must allow for changes in its
ontology so that it can be adapted for use in various environments and continually
developed in an iterative manner. As environments evolve, new types of information
appear and terminologies change.</p>
      <p>The requirement to adapt poses conflicting challenges. First, maintenance and
evolution of a structured schema must be managed. A user faced with a new
classification of information must be able to define the new entity in the schema.
Second, this process must be managed collaboratively. Designated stewards of the
ontology in a user organization must be able to make necessary modifications without
having to resort to calling for technical support.</p>
    </sec>
    <sec id="sec-2">
      <title>Approach</title>
      <p>
        Harnessing the collective intelligence of an application’s user community is an
effective way to address challenges [
        <xref ref-type="bibr" rid="ref2">2</xref>
        ]. This is exactly how a web-based semantic
wiki tackles the problem of ontology management. The approach recognizes that the
end user has the expertise to define the granular descriptive characteristics of data,
and empowers the user to do so directly through the wiki interface.
      </p>
      <p>Using initial input from the targeted user community, the developers of a
semantic wiki determine the structure of the ontology and an initial baseline lexicon
of options. Once the software is deployed, the user interface allows the operator to
create new sub-types in the supporting schema, and populates these new sub-types to
the displays of all other users. All users are instantly able to take advantage of the
modifications made by the designated data stewards, resulting in the collaborative
tailoring of mission-specific schema to meet unique requirements.</p>
      <p>Application developers carefully consider two aspects of schema management
when designing an interactive, collaborative capability: control and visibility. These
complementary properties help prevent an ontology from becoming fragmented and
losing functionality. Control is established by level and by action. Designers
determine the level of the structure—where in the hierarchy tree—at which a user can
modify the schema. Designers also determine the specific actions that the user can
take (Create new? Rename? Delete?). Visibility is maintained by repopulating the
display with changes, by providing users with a read-only view of the entire schema,
and by a recommendation engine. Visibility of the full range of existing options helps
schema control by limiting the creation of duplicative sub-types.
4</p>
    </sec>
    <sec id="sec-3">
      <title>Demonstration</title>
      <p>The demonstration uses an ontology developed under the Information Networking for
Operational Reporting and Monitoring (INFORM) project, designed to facilitate
collaborative analysis and work flows for elements of the U.S. Marine Corps, the U.S.
Special Operations Command, and the Department of State. The demonstration uses a
semantic wiki to process data from disparate sources and semantically link it based on
the ontology. Due to the wide range of environments and use cases, such as
humancentric analysis supporting Psychological Operations, the application assists the user
in ontology management by offering a certain level of control and visibility.</p>
    </sec>
  </body>
  <back>
    <ref-list>
      <ref id="ref1">
        <mixed-citation>
          1.
          <string-name>
            <surname>Hughes</surname>
            ,
            <given-names>T.</given-names>
          </string-name>
          :
          <article-title>Toward Semantic Integration of Data in the Wild</article-title>
          .
          <article-title>Invited talk</article-title>
          ,
          <source>Ontology for the Intelligence Community</source>
          <year>2007</year>
          , Columbia, Maryland (
          <year>2007</year>
          )
        </mixed-citation>
      </ref>
      <ref id="ref2">
        <mixed-citation>
          2.
          <string-name>
            <surname>Alag</surname>
            ,
            <given-names>S.</given-names>
          </string-name>
          : Collective Intelligence in Action. Manning, Greenwich, Connecticut (
          <year>2009</year>
          )
        </mixed-citation>
      </ref>
    </ref-list>
  </back>
</article>