Defense Advanced Research Projects Agency (DARPA), Contract Management Directorate (CMD), 3701 N. Fairfax Dr., Arlington, VA 22203-1714

A -- RAPID KNOWLEDGE FORMATION SOL 99-35 DUE 102599 POC Mr. Murray Burke Fax (703) 696-2203 RAPID KNOWLEDGE FORMATION (RKF). SOL BAA 99-35, due 25 October 1999, constitutes a Broad Agency Announcement as contemplated in FAR 6.102 (d) (2) (i). POC: Mr. Murray Burke, Defense Advanced Research Projects Agency (DARPA) ISO 3701 N. Fairfax Drive, Arlington, VA 22203-1714. Administrative FAX: (703) 696-2203.PROGRAM GOALS/SCOPE: The Defense Advanced Research Projects Agency (DARPA) Information Systems Office (ISO) is soliciting proposals for an on-going research effort that investigates, develops, and demonstrates technologies for the rapid development and application of large knowledge bases under a program called Rapid Knowledge Formation (RKF). This program's central objective is to enable distributed teams of subject matter experts (SMEs) to enter and modify knowledge directly and easily, without the need for specialized training in knowledge representation, acquisition, or manipulation. The resulting knowledge bases (KBs) will be available to provide specific answers to questions and could be applied in many different problem-solving situations. This technology will permit scientific, technical, and military experts to encode massive amounts of knowledge into reusable knowledge bases for application in many different tasks. The Government plans to award multiple technology development contracts in the following categories: (1) integrated teams of technology developers, (2) innovative component technology developers, and (3) challenge problem development and evaluation management. Proposals must stress the offeror's current and planned capability, and past record of accomplishment. The RKF Program is a follow-on to the High Performance Knowledge Bases (HPKB) Program. Offerors are strongly encouraged to review the results of the HPKB Program and the RKF Program briefings, which are accessible at http://www.darpa.mil/baa/#iso. Offerors must obtain the Proposer Information Packet (PIP), which is also accessible at http://www.darpa.mil/baa#iso. TECHNICAL AREAS OF INTEREST: The four technology areas are believed to be promising for attaining program objectives: (1) Human-Knowledge Base (KB) Interaction, (2) Knowledge Formation (3) Theory Manipulation, and (4) Knowledge Content. Although DARPA sees special merit in exploiting synergies among large general-purpose knowledge, analogical reasoning, and natural language dialogue techniques, offerors are free to innovate and are not limited to bidding exactly these technology areas. Equal consideration will be given to all approaches that enable SME authoring of KBs. These technology areas are discussed in detail in the Proposer Information Package (PIP). DARPA is strongly committed to a sound empirical foundation for this program based on operational scale problems. The selected challenge problem is from the general area of biological weapons acquisition and associated intelligence analysis. This is a long-standing concern to the Department of Defense. The challenge problem, will be used to provide the framework to formally evaluate the capabilities of candidate technologies, and is structured in such a way that no single knowledge base technology alone will solve the problems. It is important that teams of developers collaborate in order to produce a solution. Evaluations will take place annually for periods of 6 weeks to validate progress, with (1) metrics collected that relate to knowledge base quality such as competency, size, and reusability, and (2) metrics that relate to efficiency of KB creation and modification. The latter includes metrics such as the speed and success rate for various KB-building and editing tasks as performed by various types of users such as programmers, logicians, and especially SMEs. Two categories of knowledge entry for the challenge problem are envisioned. The first category is "textbook knowledge," that relates to standard textbooks in undergraduate biology. SMEs are graduate student teaching assistants well qualified with respect to the selected textbooks. The SME task is to teach the textbook material to the knowledge base. The second category is "expert knowledge" that relates practical, hands-on tasks for intelligence analysis regarding biological warfare (BW) development programs. SMEs are analysts experienced in developing and reasoning with models of extant or emerging BW programs. Their job is to teach the performance of the task to the KBs. Both categories will be closely linked in the BW subject domain. The challenge problems for developers will be framed in terms aligned with the pre-existing HPKB upper ontology and represented in the MELD language, which is a version of predicate calculus; a description and specification for MELD may be found at http://www.cyc.com/rkf/index.html. All answers are to be provided as MELD expressions and aligned with (using the terms from) the Integrated Knowledge Base (IKB), which was developed during HPKB, and which will be extended as required for and during RKF. We strongly advocate that all developers use the IKB and its associated interfaces/knowledge server/KR system called IDE for "Integrated Development Environment". If not, developers are required to show in their proposal how translation to/from IKB will be accomplished. Developers not using the IKB and IDE will be responsible for all translation activities and software. Limited-use IKB/IDE licenses will be provided GFE; developers must accept terms of such licenses. A sample IKB/IDE license that was used for and during HPKB, and which will be similar to the RKF license, is available online as http://www.cyc.com/rkf/index.html.PROGRAM IMPLEMENTATION: RKF envisions the following program milestones for implementation and development: Year 1: Technology development with individual component evaluation experiments for the end of the year. Year 2: Implement and test a single-subject matter expert knowledge authoring system. Year 3: If the results of Year 2 successfully demonstrates a single-user knowledge entry capability, implement and test parallel knowledge entry technology capable of supporting 25 to 50 individuals working asynchronously. Year 4: Implement and test a very large biological weapons KB (with 1 million axioms) using teams of SMEs who are AI novices. More detail is provided in the Proposal Information Package. EVALUATIONS: Each Category 1 and 2 program participant will be required to provide his/her technology to the Category 3 Evaluation Management contractor for rigorous experimentation and objective assessment by the evaluation contractor. Teams of SMEs and knowledge engineering experts will be part of the evaluation contractor's team. The objective of these assessments is to validate developer claims concerning SME usability and utility. SMEs will use the supplied technology to independently develop knowledge bases and problem solving tools. Measurements will be made concerning KB quality and SME performance. EVALUATION CRITERIA: Evaluation of proposals will be accomplished through a technical review of each proposal using the following criteria, which are listed in descending order of relative importance: Category 1: IntegrationDevelopment Teams Technical approach including synergy between component technologies Innovation associated with proposed solutions' capabilities to meet program objectives Personnel in primary and subcontractor organizations Capabilities and related experience Cost realism or best value Category 2: Component Technology Developers Dramatic innovation associated with proposed technology to support program objectives Technical approach and program plan Personnel in primary and subcontractor organizations Capabilities and related experience Cost realism or best value Category 3: Evaluation Management (1) Personnel in primary and subcontractor organizations Innovation of proposed solutions to meet program objectives Technical approach and program plan Capabilities and related experience Cost realism or best value. ADDITIONAL CONSIDERATIONS: Offerors should identify the specific area(s) they are addressing. Offerors should describe the requirements of the area from their perspective, describe the key technical challenges, and identify why the area(s) is a challenge. Offerors should describe their approach and indicate why they will be successful, particularly if other approaches have not been attempted. GENERAL INFORMATION. Final proposals are due by 4:00 PM, Daylight Savings Time on 25 October 1999 in order to be considered. Proposers must submit an original and 5 copies of full proposals, and an electronic copy on a floppy (in WordPerfect or Word for IBM compatible, MacIntosh or ASCII text format on a 3-1/2 inch floppy disk and referencing BAA 99-35), to DARPA/ISO, 3701 North Fairfax Drive, Arlington, VA 22203-1714, (ATTN: BAA 99-35). Proposers must obtain BAA 99-35 Proposer Information Package (PIP) that provides further information on areas of interest, the submission, evaluation, funding processes, and proposal formats. The PIP may be obtained by electronic mail, world-wide web, or mail request to the administrative contact address given below, as well as at URL address http://www.darpa.mil/baa#iso. This Commerce Business Daily notice, in conjunction with the Proposer Information Package, constitutes the total BAA. Proposals not meeting the format described in the package may not be reviewed. Proposals submitted by fax, will be disregarded. No additional information is available, nor will a formal RFP or other solicitation regarding this announcement be issued. Requests for same will be disregarded. The Government reserves the right to select for award all, some, or none of the proposals received. All responsible sources capable of satisfying the Government's needs may submit a proposal for consideration. Historically 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 will be set aside for HBCU and MI participation due to the impracticality of reserving discrete or several areas of research in this area of research. ORGANIZATIONAL CONFLICT OF INTEREST: Awards made under this BAA are subject to the provisions of the Federal Acquisition Regulation (FAR) Subpart 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 should be furnished at the time of proposal submission. All facts relevant to 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 Contractor has taken, or proposes to take, to avoid, neutralize or mitigate such conflict. It is the policy of DARPA to treat all proposals as competitive information and to disclose the contents only for the purposes of evaluation. The Government intends to use Schafer Corporation personnel as special resources to assist with the logistics of administering proposal evaluation and to provide advice on specific technical areas. Personnel of this contractor 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 Organizational Conflict of Interest Non-Disclosure Agreements (OCI/NDA). By submission of its proposal, each offeror agrees that proposal information may be disclosed to this selected contractor for the limited purpose stated above. Any information not intended for limited release to this contractor must be clearly marked and segregated from other submitted proposal material. ADMINISTRATIVE ADDRESSES: All administrative correspondence and questions on this solicitation, including requests for information on how to submit a proposal to this BAA, should be directed to DARPA/ISO, ATTN: BAA 99-35, 3701 N. Fairfax Drive, Arlington, VA 22203-1714. Email or fax is preferred. Electronic Mail: baa99-35@darpa.mil, Electronic File Retrieval: http://www.darpa.mil/baa/#iso, Fax: (703) 516-6065. DARPA intends to use electronic mail and fax for correspondence regarding BAA 99-35.SPONSOR: Defense Advanced Research Projects Agency (DARPA), Contract Management Directorate (CMD), 3701 N. Fairfax Dr., Arlington, VA 22203-1714 SUBFILE: PSE (U.S. GOVERNMENT PROCUREMENTS, SERVICES) SECTION HEADING: A Research and Development PUBLICATION DATE: ISSUE: PSA- Posted 08/31/99 (W-SN374946). (0243)

Loren Data Corp. http://www.ld.com (SYN# 0010 19990902\A-0010.SOL)