SOURCES SOUGHT
70 -- Joint Intelligence Virtual Architecture Program (JIVA)
- Notice Date
- 1/15/2002
- Notice Type
- Sources Sought
- Contracting Office
- Other Defense Agencies, Defense Intelligence Agency, Virginia Contracting Activity (ZD50), 200 MacDill Boulevard Post Office Box 46563, Washington, DC, 20035-6563
- ZIP Code
- 20035-6563
- Solicitation Number
- Reference-Number-MDA908-02-Q-0028
- Response Due
- 2/15/2002
- Archive Date
- 3/2/2002
- Point of Contact
- James Dashiell, Branch Chief - CO, Phone (202) 231-2947, Fax (202) 231-2831, - Ida Batson, Contracting Officer, Phone (202) 231-8290, Fax (202) 231-2831,
- E-Mail Address
-
jim.dashiell@dia.mil, ida.batson@dia.mil
- Description
- The Virginia Contracting Activity (VaCA), on behalf of The Defense Intelligence Agency (DIA), and the Joint Intelligence Virtual Architecture (JIVA) program is seeking white papers for a highly integrated Knowledge Discovery (KD) toolkit. This is a sources sought announcement. JIVA is a congressionally mandated program within the General Defense Intelligence Program (GDIP) and is the primary DOD intelligence community?s program for infusion of commercial information technology in support of intelligence analysis and production processes. Because the desired suite of KD tools does not presently exist as an integrated set, development of the KD toolkit will likely be a technical and functional integration of emerging state of the art components with each other and within the JIVA Enterprise architectural framework. White papers are being sought from both Industry and Government sources, and should specify a complete end-to-end solution that satisfies the top-level requirements identified below. The immediate goal is to identify candidate KD tools to evaluate as part of a JIVA lab-based concept exploration effort during the summer of 2002, with accreditation and deployment on the operational JIVA Enterprise in the first quarter of Fiscal Year 2003. The KD toolkit is required to support immediate on-going DoD Intelligence Counter Terrorism efforts. However, its use will likely extend beyond Counter Terrorism to the general military intelligence analyst population and the JIVA enterprise. Traditional intelligence problems, such as assessing the military forces, capabilities, and intentions of foreign nations, generally involve analyzing data within both structured and unstructured databases and known hierarchies. However, the asymmetric threat problem, specifically the terrorism threat, involves assimilating a great deal of data from the CIA, NSA, DoS, FBI and INS; Intelligence Community real time message feeds and mail systems; and open source HTML/XML and free text. In addition, the Counter Terrorism analyst must often think outside the box in terms of established rational Western Standards of analysis and thus the ability to conduct what-if drills is paramount. A central concept involves establishing relationships between data points and testing operational assumptions. Nevertheless, because of the volume and inconsistency of raw data, it is imperative that proven analytical methodologies be applied and that advanced information technology tools are employed. This KD toolkit is envisioned to provide the Counter Terrorism analyst with an end to end solution that can: (1) Create refined complex queries (searches) to manually identify/extract entities/relationships or the intelligent agents to find them automatically from the JIVA Virtual Knowledge Base (VKB), and potentially other compartmented data sources/repositories used by the CT community. The evolving VKB contains all types of unclassified and classified multi-media, unstructured, semi-structured, and structured data. VKB will provide seamless access to intelligence counter terrorism data bases and other databases such as MIDB, MEPED, EWIR, and focus on global data access. It will be important for the tools to filter out unnecessary data, process incoming data in near real time, apply analyst defined taxonomies, and store data in an ODBC/JDBC compliant database. Analysts will need to be able to query sources using both natural language and key words. This component is called the Advanced Data Mining component; (2) Visually recognize significant patterns, relationships, and timelines in those returned results. It will be important for the analyst to exploit the data for further discovery. This component is called the Data Visualization component; (3) Create and test hypotheses through a question and answer interface that can resolve ambiguities and enable the analyst to build and test what if scenarios. This component is called the Question and Answer component; and, (4) Build the final intelligence analysis through use of a Link Discovery and Visualization tool that can structure output based on user pre-defined relationships (models, formats, templates). Through this component, the analyst can visualize and present the linkages uncovered in the data for use by other analysts throughout the community, overlay the analysis on imagery or maps; and discover unknown relationships based on automated natural language processing. The JIVA program has already invested in several potentially relevant technologies that can be applied to the intelligence analysis problem and consideration should be given to exploiting the capabilities of these tools. If recommended, they can be integrated into the KD toolkit. These are Visual Analytic?s Visual Links, Autonomy, Presearch?s Pathfinder for the Web analytical tools, and the JIVA Geographic Information System which is built around ESRI?s ARCIMS software. It is strongly emphasized that white papers will not be penalized in any way if they do not include any of these tools. JIVA is forward-looking and is not wedded to any of its existing capabilities; however, JIVA is hopeful that some of its past investments can be exploited in an integrated KD solution. The proposed Knowledge Discovery Toolkit must be compatible with the JIVA three-tier (Presentation (web browser), Business/Application logic, Data (VKB)) enterprise software architecture. As such it must meet the following critical requirements: (1) scalable to a large geographically separated analyst population numbering in the thousands on the JWICS SCI network with possible extension to the SIPRnet SECRET environment.; (2) Web-enabled without/minimal client load, IE/Netscape independent, and Java Run-Time Environment (JRE) 1.3 compatible; (3) capable of being accessed and secured through the Sybase Enterprise Portal 2.0 framework which is the implementation of the JIVA Enterprise software architecture; and, (4) supportive of a federated instantiation of geographically separated clustered servers managed by a combination of the JIVA Enterprise Management System (JEMS) software composed of Tivoli and BMC modules, and hardware by F5. Full J2EE compliance is strongly desired to ease integration within the JIVA Enterprise software architectural framework (Sybase EP 2.0). A major component of the back end data repositories is in IBMs DB2 format. The analytical review of submitted white papers will heavily emphasize the following in priority order: (1) Potential effectiveness in providing focused responses to analyst queries, and, ability to discover new relevant information; (2) Simplicity of interface and ease of use without training; and, (3) Compatibility with a web-based enterprise architecture and scalability to a large analyst population. Responses should be unclassified, as short as possible, but in no case more than ten 8.5x11 pages. Responses are requested 20 business weekdays from the date this announcement is posted. Responses should be provided electronically via an e-mail enclosure (in MS Word 2000 format without macros) to the following addresses: mdshaffer@hotmail.com; dennis.lair@dia.mil; jim.dashiell@dia.mil. The file should be uncompressed and not exceed 200 KB in size. Any amplifying information (e.g. white papers) should be posted on the Internet and URL pointers provided within the response. Questions regarding this RFI should be directed to either Mike Shaffer (202)231-2189 or Dennis Lair (202)231-3821. Upon completion of the white paper review, the JIVA Program Office will select up to three of the most promising solutions presented and request briefings at the Defense Intelligence Analysis Center, Bolling AFB, Washington, D.C. If merited, a contract may be extended to up to three sources to build and demonstrate a funded concept exploration prototype in the JIVA Secure Test and Integration Center (STIC). This prototype will be connected to real-world SCI data and be evaluated by a group of analysts and technical experts. It is highly desirable to have SCI cleared development personnel available to support the initial installation and connection to data sources and analyst training. The concept exploration prototypes must be ready for analyst evaluation by 1 June 2002 because the JIVA program intends to commit to an enterprise deployment decision before the end of September 2002. The person assigned this sources sought is Mr. James Sparks (202) 231-2134.
- Place of Performance
- Address: Defense Intelligence Agency, Bldg. 6000, Bolling AFB, Washington, DC
- Zip Code: 10230
- Country: USA
- Zip Code: 10230
- Record
- SN20020117/00014970-020116090840 (fbodaily.com)
- Source
-
FedBizOpps.gov Link to This Notice
(will not be valid after Archive Date)
| FSG Index | This Issue's Index | Today's FBO Daily Index Page |