COMMERCE BUSINESS DAILY ISSUE OF FEBRUARY 27, 2001 PSA #2796
SOLICITATIONS
A -- EVIDENCE EXTRACTION AND LINK DISCOVERY (EELD) PROGRAM
- Notice Date
- February 23, 2001
- Contracting Office
- Defense Advanced Research Projects Agency (DARPA), Contract Management Office (CMO), 3701 N. Fairfax Dr., Arlington, VA 22203-1714
- ZIP Code
- 22203-1714
- Solicitation Number
- BAA 01-27
- Response Due
- January 31, 2002
- Point of Contact
- Alan Frederick, Contracts Management Office, DARPA/CMO, FAX (703) 696-2208
- E-Mail Address
- Click here for administrative, technical or contractual (baa01-27@darpa.mil)
- Description
- The Defense Advanced Research Projects Agency (DARPA) Information Systems Office (ISO) is issuing this single step (full-up proposals, no white papers) Broad Agency Announcement (BAA) to solicit research proposals (structured as a 30 month base period plus two (2) one (1) year options) for the EELD Program. The goal of this program is to develop, demonstrate, and evaluate technology components in three areas necessary to enable constructing EELD systems: Evidence Extraction, Link Discovery, and Pattern Learning. It also includes work in the area of system concept and performance analysis, the goals of which are to provide a system and user perspective, and to measure the increase in capabilities of the technology developments. The Government anticipates multiple awards in each of three technical areas; single awards, perhaps to teams of organizations, in the system concept and the performance analysis areas. RESEARCH NEED: A serious threat to US national security today is asymmetric; i.e., not the threat of a conventional force-on-force engagement by an opposing military, but the threat of an unconventional yet highly lethal attack by a loosely organized terrorist group. Detection of these asymmetric threats requires a new level of autonomic information surveillance and analysis to extract, discover, and link together sparse evidence from vast amounts of classified and unclassified data sources (especially open sources), to represent and evaluate the significance of the related evidence, and to learn patterns to guide the extraction, discovery, linkage, and evaluation process. The theme of the EELD program is the extension of technical capabilities in the three technical areas of interest to structured, or relational, data. These relationships could be transactional, social, temporal, or geographical. Of particular interest is the extraction and discovery of related temporal events that may be components of scenarios of interest and the learning of patterns that comprise such scenarios. Evidence extraction technology will be extended from its current ability to extract accurately named entities and attributes to the ability to extract relationships between entities and attributes of these relationships. Link discovery technology will be developed to enable the ability to discover related entities, additional attributes, and other relevant relationships from a starting point of a set of entities, relationships, and attributes potentially relevant to a scenario of interest. Pattern learning technology will be developed to enable learning from examples of instances of scenarios of interest patterns or models which will facilitate the extraction and discovery of additional instances of scenarios of interest. This announcement constitutes a Broad Agency Announcement (BAA 01-27) in accordance with the Federal Acquisition Regulation (FAR) at 6.102(d)(2)(i). Information in this announcement is NOT complete. Prospective offerors MUST also consult the Proposer Information Pamphlet (PIP) for the EELD Program before submitting a proposal. Long Range Vision: The vision for the EELD Program is to advance significantly the state-of-the-art in Evidence Extraction (EE), Link Discovery (LD), and Pattern Learning (PL) technology to enable the construction of prototype systems that can be used for asymmetric threat detection. Such a system would extract entities, relationships, and their attributes from textual or semi-structured data sources, integrate facts from structured databases, create and update profiles of entities and scenarios of events, link together related entities and events, and evaluate these linked structures for activities of interest. For interesting instances it would discover additional related entities and attributes to enable a determination as to whether or not to issue an alert. It would also be able to learn, from small numbers of examples, models of scenarios of interest, to be used to guide the evidence extraction and link discovery processes. PROGRAM SCOPE: The overall approach to developing, demonstrating, and evaluating the EE, LD, and PL technology is: (1) analyze the EE, LD, and PL needs of the intelligence analysts, and refine the EELD system concept to reflect these needs; (2) develop a set of unclassified challenge problems and associated data sets and domain models/knowledge bases appropriate for testing and evaluating advances in each technical area, and sharing characteristics with the actual asymmetric threat detection problem; (3) research and develop promising approaches to distinct aspects of evidence extraction to advance significantly the state-of-the-art; (4) research and develop promising approaches to distinct aspects of link discovery to advance significantly the state-of-the-art; (5) research and develop promising approaches to distinct aspects of pattern learning to advance significantly the state-of-the-art; (6) evaluate these approaches using the unclassified challenge problems; (7) as an area matures, integrate various techniques and approaches (with distinct strengths and weaknesses) that address multiple data types and requirements aspects within the area into a single component that provides the level of capability required for the construction of an effective EELD system; and (8) develop classified asymmetric threat challenge problems and conduct evaluations of selected integrated EE, LD, and PL components on these classified challenge problems. The program plan will consist of parallel technology development efforts in each of the three technical areas. Annual evaluations in each of these areas will be conducted against a series of challenge problems. TECHNICAL TOPIC AREAS: There are three primary technology areas in which advances are necessary, and one supporting area in which work is needed, to achieve program goals. These areas and their research foci are: 1. Evidence Extraction. Approaches for extracting relational facts. 2. Link Discovery: Approaches for discovering relevant linkages between entities, reasoning about different levels of abstraction, classifying the emerging and suspected instances of patterns of interest, and reasoning with multiple hypotheses and with uncertainty. 3. Pattern Learning. Approaches for enabling a system to learn, from a small number of example instances consisting of data about entities, relationships, and their attributes, models of scenarios of interest. 4. System Concept and Performance Analysis. User and system perspectives relative to the technology developments in the three primary technical areas; the development of challenge problems, and the evaluation of the technology developments. PROGRAM IMPLEMENTATION: A multi-phase approach will be needed to achieve program goals. The degree of advances will vary among the three technical areas of interest. Current EE capabilities are closest to needed performance, and EE is considered the most mature of the three areas. PL capabilities are the least mature, and the achievement of desired performance in this area will require the greatest advance in capability. The maturity of LD falls between the other two areas. Because of these different levels of technological maturity, DARPA anticipates awarding a smaller number of efforts in the EE area, a larger number of efforts in the PL area, and an appropriate number of efforts in the LD area. Offerors should note that approaches that combine ideas from distinct technical specialties or communities are actively sought. ORGANIZATIONAL CONFLICT OF INTEREST: 1. General Requirements. Each cost proposal shall contain a section satisfying the following requirements: awards made under this BAA are subject to the FAR at 9.5, Organizational Conflict of Interest. All offerors and proposed subcontractors must affirmatively state whether they are supporting any DARPA technical office(s) through an active contract or subcontract. All affirmations must state which office(s) the offeror supports and identify the prime contract number. Affirmations shall be furnished at the time of proposal submission, and the existence or potential existence of organizational conflicts of interest, as that term is defined in FAR 9.501, must be disclosed. This disclosure shall include a description of the action the offeror has taken, or proposes to take, to avoid, neutralize or mitigate such conflict. If the offeror believes that no such conflict exists, it shall so state in this section. It is the policy of DARPA to treat all proposals as competitive information, and to disclose the contents only for the purposes of evaluation. 2. Special Resource Personnel. The Government intends to use personnel from Schafer Corporation of Arlington, Virginia as special resources to assist with the logistics of administering proposal evaluation, and to provide advice on specific technical areas. These contractor personnel are restricted by their contracts from disclosing proposal information for any purpose other than these administrative or advisory tasks. Contractor personnel are required to sign the Organizational Conflict of Interest Non-Disclosure Agreements (OCI/NDA). By submission of its proposal, each offeror agrees that proposal information may be disclosed to these selected contractors for the limited purpose stated above. Any information not intended for limited release to these contractors must be clearly marked and segregated from other submitted proposal material. 3. Additional Information. DARPA is seeking participation from the widest number of offerors. Toward this end, DARPA is willing to consider all types of proposals including traditional FAR/DFARS type contracts, grants, and "Other Transactions." Offerors interested in pursuing an "Other Transaction" under this BAA should contact Alan Frederick, Contracts Management Office (CMO), DARPA at 703-696-0047 or afrederick@darpa.mil. EVALUATION CRITERIA: Proposals, properly formatted and delivered in accordance with this BAA, shall be evaluated pursuant to four criteria. These criteria, in their order of priority, are: (1) Technical Approach including innovation, validation, potential contribution, and system and user considerations; (2) Approach to Technology Transfer; (3) Offeror's Capabilities and Related Experience and (4) Cost Realism. The PIP provides further information concerning these criteria. PROPOSAL SUBMISSION: Proposers MUST consult the BAA 01-27 PIP, which includes further information on the areas of interest, such as technical overview, program schedule, expected deliverables, preparation and formats for proposals, and proposal evaluation criteria. The PIP may be accessed via the World Wide Web one working day after release of the BAA in the CBD publication at URL http://www.darpa.mil (search under "Solicitations"), or requests may be sent via e-mail to baa01-27@darpa.mil. To be considered for the initial round of funding, full proposals must be submitted to DARPA/ISO, 3701 North Fairfax Drive, Arlington, VA 22203-1714 (Attn: BAA 01-27) on or before 4:00 p.m., local time, Wednesday, 4 April 2001. In addition to hardcopies, electronic copies of proposals should be submitted on floppy diskette(s) or Iomega 100MB Zip disk(s) according to instructions in the PIP. Proposals submitted by fax or e-mail are not acceptable and will not be considered. The Government anticipates completing evaluations during the third quarter of fiscal year 2001. PROPOSAL FORMAT: Proposals shall consist of two separately bound volumes. Volume I shall provide the technical proposal and management approach and Volume II shall address the cost. Detailed instructions for completing the proposals are contained in the PIP. Proposals not meeting the specified format described in the PIP will not be reviewed. OTHER GENERAL AND ADMINISTRATIVE REQUIREMENTS: DARPA will construct a balanced program in order to meet its needs. A total of approximately $2.7M will be available in FY01 for multiple efforts. In FY02, there is approximately $9M anticipated. In FY03 there is approximately $10M. In FY04 there is approximately $11M and in FY05 about $6.5M is expected. Size of each award and duration of efforts will vary according to the type of effort undertaken. All responsible sources may submit a proposal. Historical Black Colleges and Universities (HBCU) and Minority Institutions (MI) are encouraged to submit proposals and join others in submitting proposals; however, no portion of this BAA 01-27 will be set aside for HBCU and MI participation due to the impracticality of reserving discrete or severable areas of technology for exclusive competition among these entities. The Government reserves the right to select for award all, some or none of the proposals received in response to this announcement. This BAA and its PIP comprise the entire solicitation for the EELD Program. COMMUNICATING WITH DARPA: All administrative correspondence and questions concerning this BAA must be directed, in writing, to one of the administrative addresses below. Electronic mail (e-mail) is preferred. DARPA intends to use electronic mail for most technical and administrative correspondence regarding this BAA. Technical and contractual questions should include the originator's full name and return e-mail address in the text. All questions and answers will be posted to URL http://www.darpa.mil (search under "Solicitations"). ADMINISTRATIVE ADDRESSES: The PIP and Frequently Asked Questions (FAQ) are available at URL http://www.darpa.mil (search under "Solicitations"). Written requests for these documents may be sent by facsimile to 703-516-6065 addressed to ATTENTION BAA 01-27 INFORMATION, or by surface mail addressed to DARPA/ISO, ATTN: BAA 01-17 INFORMATION, 3701 N. Fairfax Drive, Arlington, VA 22203-1714. Administrative, technical or contractual questions should be sent via e-mail, if possible, to baa01-27@darpa.mil. If e-mail is not available, please direct questions to one of the above addresses. These requests must include the name, address, and phone number of a point of contact. Posted 07/XX/00 (W-SN3408XX). (0160)
- Web Link
- Visit this URL for the latest information about this (http://www.darpa.mil )
- Record
- Loren Data Corp. 20010227/ASOL001.HTM (W-054 SN50E4F8)
| A - Research and Development Index
|
Issue Index |
Created on February 23, 2001 by Loren Data Corp. --
info@ld.com
|
|
|
|