SOLICITATION NOTICE
R -- Automatic Generation of a Biomedical Knowledge Base for Semantic Search and Semantic Web Applications
- Notice Date
- 10/15/2010
- Notice Type
- Combined Synopsis/Solicitation
- NAICS
- 541990
— All Other Professional, Scientific, and Technical Services
- Contracting Office
- Department of Health and Human Services, National Institutes of Health, National Library of Medicine, 6707 Democracy Blvd., Suite 105, Bethesda, Maryland, 20894, United States
- ZIP Code
- 20894
- Solicitation Number
- HHS-NIH-NLM-11-002-SES
- Archive Date
- 11/14/2010
- Point of Contact
- Shari E. Shor, Phone: 301-435-4388, Robin D Hope-Williams, Phone: 301-435-4379
- E-Mail Address
-
shorse@mail.nlm.nih.gov, rhwilli@mail.nih.gov
(shorse@mail.nlm.nih.gov, rhwilli@mail.nih.gov)
- Small Business Set-Aside
- Total Small Business
- Description
- This is a synopsis/solicitation for commercial services prepared with the format under Simplified Acquisition Procedures at Federal Acquisition Regulation (FAR) Subpart 12.6 Streamlined Procedures for Evaluation and Solicitation for Commercial Items; and FAR Subpart 13.5, Test Program for Certain Commercial Items; as supplemented with additional information included in this notice. This notice constitutes the only solicitation; proposals are being requested and a written solicitation will not be issued. This solicitation is being issued as a Request for Quotations (RFQ) HHS-NIH-NLM-11-002-SES. This solicitation document incorporates provisions and clauses that are in effect in the April 2009 Federal Acquisition Regulation (FAR) Revision, including all FAR Circulars issued as of the date of this synopsis. This acquisition is 100% small business set-aside and the North American Industry Classification System (NAICS) code is 541990, All Other Professional, Scientific, and Technical Services. Background The National Library of Medicine (NLM), the Specialized Information Services Division (SIS), has developed innovative technologies to provide information services to the public in toxicology and environmental health, HIV/AIDS, chemical information and disaster information management. Applied research and development has played a vital role in the development and implementation of new tools, databases and search services in the Division. Project Objectives The purpose of this procurement is to architect and deploy innovative natural language processing (NLP) and information retrieval (IR) solutions for the automatic generation and continued enhancement of a large Biomedical Knowledge Base (BKB) to be used in the implementation of semantic search engines and semantic web applications to support of the Division's mission. These new capabilities are expected to enable SIS to offer better quality search results and information syntheses for NLM's diverse user communities, from ordinary citizens who seek reliable information related to health concerns, to scientists on the cutting edge of biomedical research and new discoveries. Scope of Work The contractor is required to perform the following Tasks: 1. Develop NLP software o Implement state-of-the-art NLP tools to identify biomedical concepts and their relationships in unstructured biomedical texts 2. Implement advanced text mining, data fusion and knowledge base (TDKB) applications for heterogeneous biomedical databases o Analyze, distill and synthesize the content of a broad spectrum of trusted biomedical information sources 3. Create a large Biomedical Knowledge Base o use the NLP tools specified in Task 1, and the TDKB applications specified in Task 2, to automatically generate a comprehensive Biomedical Knowledge Base o Utilize commonly accepted knowledge representation and dissemination standards, for instance XML, RDF and semantic triples for the BKB 4. Implement, test and deploy a Web Service for the searching and utilization of the Biomedical Knowledge Base specified in Task 3 5. Deliver and install the NLP tools, knowledge base applications, the BKB and the BKB Web Service software specified in Tasks 1 through 4 on an SIS server Detailed Technical Requirements The detailed technical requirements for each of the tasks are as follows: 1. (Task 1) Develop NLP software o The contractor is required to implement state-of-the-art NLP tools to identify biomedical concepts and their relationships in unstructured biomedical texts  The NLP tools must be able to process biomedical texts, including full text journal articles, typically found in high quality biomedical databases and consumer health web sites, including but not limited to PubMed, ClinicalTrials.gov, Medlineplus and the content of the various National Institutes of Health web sites  The NLP tools must be able to explicitly identify specific biomedical entities of interests including, but not limited to • Diseases and Conditions • Signs and Symptoms • Tests and Diagnoses • Treatments and Procedures • Drugs and Substances • Genes • Biomarkers • Complementary and Alternative Medicine Approaches • Definitions of the entities of interest  The NLP tools must be able to identify the nature and contexts of the relationships between the entities of interest, including but not limited to • human-readable text that provides a clear and concise context that reflects the nature of the given relationship between entity pairs of interest • a measure of the strength of association between entities of interest  The NLP tools must include but need not be limited to the following specific tools: • TAGGER - a part of speech tagger for biomedical English text • PHRASER - a parser to identify noun phrases and verb phrases in biomedical English text • SPELL - a spelling detection and correction system for biomedical English text 2. (Task 2) The contractor is required to implement advanced text mining, data fusion and knowledge base applications for the content processing of heterogeneous biomedical databases  The TDKB applications to be developed in this task must integrate and utilize the NLP tools developed in Task 1 in order to analyze, aggregate and synthesize content from the heterogeneous biomedical databases specified in Task 1. More specifically, • the TDKB applications to be developed in this task must be able to identify and merge the entities and relationships implicitly or explicitly present in the content of high quality biomedical databases and consumer health web sites listed in Task 1 • the TDKB applications to be developed in this task must be able to integrate the merged biomedical content entities and entity relationships with concepts and relationships implicitly or explicitly present in NLM's Unified Medical Language System (UMLS) • the TDKB applications to be developed in this task must be able to integrate the merged biomedical content entities and entity relationships with concepts and relationships implicitly or explicitly present in other publicly available ontologies and knowledge sources such as Wikipedia and Freebase • The TDKB software applications to be developed in this task must include but need not be limited to the following specific components: o TEXT2UMLS - an application to contextually map tokens and phrases in biomedical texts to UMLS Metathesaurus concepts and semantic type codes 3. (Task 3) Create the Biomedical Knowledge Base • utilizing the software and data created in Tasks 1 and 2, the contractor must automatically generate a comprehensive Biomedical Knowledge Base • the Biomedical Knowledge Base must minimally include, but need not be limited to the entities and relationships of interest specified in Tasks 1 and 2 above • The Biomedical Knowledge Base must be minimally comprised of, but need not be limited to records that include all unique concepts in Release Version 2010AA (or later versions) of the UMLS Metathesaurus. The minimum size of the BKB is estimated to be 3 million unique concept records. • The Biomedical Knowledge Base must minimally comprised of, but need not be limited to records that contain the specific types of biomedical entities and their relationships specified in Tasks 1 and 2. The minimum number of entity pairs and their relationships ("semantic triples") in the BKB is estimated to be three billion. • The semantic triples within BKB records must be ranked by semantic similarity and strength of association 4. (Task 4) Implement and deploy a Web Service for the searching and utilization of the Biomedical Knowledge Base 1. The BKB Web Service must return explicit and detailed information about entities (concepts) and their relationships in the Biomedical Knowledge Base, as specified in Sections 1, 2 and 3 above 2. Specific Requirements for the BKB Web Service are: • Ability to handle the lexical/morphological, morpho-semantic, syntactic, semantic and pragmatic characteristics and problems inherent in biomedical English • Ability to handle synonyms and quasi-synonyms • Ability to handle abbreviations and acronyms • Ability to handle generic medical concept qualifiers and facets • Ability to handle complex queries • Ability to disambiguate queries • Ability to handle lay language terms • Ability to handle biomedical and scientific misspellings • 5 (Task 5) Deliver and install the NLP tools, TDKB applications, the Biomedical Knowledge Base and the BKB Web Service specified in Tasks 1-4 on an SIS Linux server, The average response time of the BKB Web Service must be under 200 milliseconds on a commodity PC Linux server Contractor Personnel The project must be conducted by contractor staff with high levels of expertise in computer science, computational linguistics, medical informatics, knowledge engineering and Java software development. The contractor must present evidence that the proposed key personnel have had instrumental roles in the design and successful production level implementations of biomedical knowledge bases and semantic search applications in biomedicine. Specifically, the proposed contractor personnel must have significant working familiarity with the following NLM biomedical information resources or similar biomedical information resources: UMLS, Pubmed and Medlineplus. In addition, the proposed contractor personnel must have significant working experience in the design and production level implementations of biomedical NLP tools, such as part-of-speech taggers, phrase parsers, medical and scientific spellcheckers and UMLS-related knowledge applications. Location The project implementation will take place at the contractor's location. Travel No travel will be required. Period of Performance It is anticipated that the period of performance shall be twelve (12) months from the date of award. An award is anticipated to be made on or about November 5, 2010. It is estimated that this will be an effort of approximately 1920 hours. Total hours worked are subject to available funding. Deliverables The key results of this project will be the NLP and TDKB tools and applications, as well as the Biomedical Knowledge Base and the Web Service for accessing the Biomedical Knowledge Base. The project is anticipated to take one year. Deliverables and Delivery Schedule The schedule for the deliverables is as follows: 1. Deliver and install the NLP tools (source code and executables) on an SIS server by the end of the 6th month from the effective date of the contract 2. Deliver and install the TDKB applications software (source code and executables) on an SIS server by the end of the 9th month from the effective date of the contract 3. Delivery and install the Biomedical Knowledge Base on an SIS server by the end of the 10th month from the effective date of the contract 4. Deliver and install the BKB Web Service software (source code and executables) on an SIS server by the end of the 11th month from the effective date of the contract The contractor is required to make brief bi-monthly progress reports and is encouraged to make available the software and data deliverables to the NLM Project Officer prior to the delivery schedule, whenever possible, so that appropriate testing and feedback is possible. Acceptance of the deliverables will be based on comparison with best-of-class NLP tools and biomedical knowledge bases within the scope of the defined tasks. Rights in Data The NLM will have unrestricted rights, for NLM purposes only, to use the software and data developed under this procurement. Statement of Impact on Other Program Areas This project does not duplicate or adversely impact any other NLM program areas. The following clauses and provisions cited herein are incorporated by reference into this solicitation and may be obtained from the web site http://rcb.cancer.gov/rcb-internet/SAP/sap.htm: FAR 52.212-1, Instructions to Offerors-Commercial (June 2008), and FAR 52.212-4, Contract Terms and Conditions-Commercial Items (June 2010). The attached Addendum to Terms and Conditions of Purchase Order also applies to this solicitation. FAR 52.212-2, Evaluation-Commercial Items (January 1999), also applies with the following four evaluation criteria to be included in paragraph (a) of the provision: Evaluation Criteria Understanding the Problem (20 Points) Proposal demonstrates a thorough and complete understanding of the requirements and indicates a clear awareness of the contract objectives. Proposal provides evidence that the offeror has a deep knowledge of the subject to be able to anticipate and avoid problems, and to react appropriately when problems do arise. Soundness of the Approach (25 Points) Proposal describes the proposed approach to comply with each of the requirements specified in the Statement of Work. The proposal is consistent with the stated goals and objectives. The proposed approach of ensuring the achievement of timely and acceptable performance is well documented and sound. Milestone and/or phasing charts illustrate the logical sequence of proposed work events, technical accomplishments and deliverables. Personnel (45 Points) 1. The proposed staff is competent and experienced in the skills required in the Statement of Work. Resumes of staff and consultants reflect not only academic qualifications, but also length and variety of experience in similar tasks and clearly demonstrate relevant training and work experience and accomplishments. If subcontractors are proposed, information is provided to support the qualifications of the subcontractors. 2. Information is provided as to which key personnel will be used on this project. Documentation is provided as to the decision-making authority of the project director as related to other elements of the organization. The percentage of time each project staff member will contribute to the program is adequately identified. The extent to which outside consultants or specialists will be used is documented and evidence of their availability is provided. Past Performance (10 Points) (Demonstrated commitment to customer satisfaction and timely delivery of high quality products and services.) a) Offerors shall submit a list and description of examples of contracts completed during the past three years and contracts currently in process. Offerors shall be evaluated on (1) record of conforming to specifications and to standards of good workmanship; (2) adherence to contract schedules, including the administrative aspects of performance; (3) reputation for reasonable and cooperative behavior and commitment to customer satisfaction; and (4) business-like concern for the interests of the customer. b) NLM will contact the references provided in order to assess the offeror's past performance and the comparability of the previous experience with NLM's stated requirements. It is the responsibility of the vendor to report non-receipt of payment to Office of Financial Management. The phone number for payment inquiries is 301-496-6088. The vendor shall keep the COTR informed about all invoicing problems. An award will be made to the offeror who represents the best value to the Government. Offerors must submit a completed copy of the provision at FAR 52.212-3, Offeror Representations and Certifications-Commercial Items (October 2010), with their offers. The clause at FAR 52.212-5, Contract Terms and Conditions Required to Implement Statutes or Executive Orders-Commercial Items (October 2010), as well as the following clauses cited therein: FAR 52.219-6, Notice of Total Small Business Set-Aside (June 2003), and FAR 52.232-33, Payment by Electronic Funds Transfer-Central Contractor Registration (October 2003). Sources having the ability to provide the professional services described above and Addendum to Terms and Conditions of Purchase Order shall submit clear comprehensive information supporting their experience, past performance, managerial experience and pricing. An offeror's response shall not exceed 10 pages, excluding resumes. Any questions can be submitted to Shari Shor, Contract Specialist, at shorse@mail.nlm.nih.gov or 301-435-4388. All information received will be considered as part of a competitive acquisition. RESPONSES ARE DUE BY 1:00 PM LOCAL PREVAILING TIME ON October 30, 2010, AND SHALL BE SENT VIA E-MAIL TO: shorse@mail.nlm.nih.gov. ALL RESPONSIBLE SOURCES MAY SUBMIT A QUOTATION WHICH, IF TIMELY RECEIVED, SHALL BE CONSIDERED BY THE GOVERNENT.
- Web Link
-
FBO.gov Permalink
(https://www.fbo.gov/spg/HHS/NIH/OAM/HHS-NIH-NLM-11-002-SES/listing.html)
- Place of Performance
- Address: Contractor location, United States
- Record
- SN02311963-W 20101017/101015234000-9fb22e521be3a4c13a1694d73d4d0e2a (fbodaily.com)
- Source
-
FedBizOpps Link to This Notice
(may not be valid after Archive Date)
| FSG Index | This Issue's Index | Today's FBO Daily Index Page |