SOLICITATION NOTICE
B -- Integrating Machine Learning Models With Business Rules Environment
- Notice Date
- 7/30/2008
- Notice Type
- Presolicitation
- NAICS
- 541690
— Other Scientific and Technical Consulting Services
- Contracting Office
- Department of the Treasury, Internal Revenue Service (IRS), National Office Procurement (OS:A:P), 6009 Oxon Hill Road, Suite 500, Oxon Hill, Maryland, 20745
- ZIP Code
- 20745
- Solicitation Number
- M-8-M0-23-OR-C17-000
- Response Due
- 8/15/2008
- Archive Date
- 8/30/2008
- Point of Contact
- Alan J Monico, Jr,, Phone: 202-283-1442, Aloysius R Hart,, Phone: 202-283-1385
- E-Mail Address
-
alan.j.monico@irs.gov, aloysius.r.hart@irs.gov
- Small Business Set-Aside
- N/A
- Description
- The Internal Revenue Service (IRS) intends to issue a sole source, firm-fixed price contract for continuation of work started in FY2007 to explore the applicability of machine learning algorithms in developing business rules to identify potential non-compliance at the tax issue level, and to estimate the validity of combining individual issues rules into a single compliance estimation model. The Contractor will assist the IRS in the development of a machine learning tool that will assist the IRS operating divisions in modernizing tax administration functions improvement on tax compliance. The issue coverage for this continuation effort will include all issues that have sufficient tax adjustment data gathered from the TY2001 National Research Program, using the same criteria as the initial 11 issues selected for modeling in FY07. In addition to the expanded issue set, the Contractor will be required to develop a methodology to combine the individual issue information into an overall model which could provide an indicator of the level of non-compliance for the tax return entity. The Contractor shall also provide measurement of the level of performance on the combined model using the newest NRP data from TY2006 that is being gathered currently by the IRS for Form 1040 tax returns. The Contractor must possess the following: expertise in recent advancements in hardware, software, and problem solving techniques such as machine learning algorithms and business rules applications; knowledge and experience with the design and development of data modeling methods and subsequent performance measurement of taxpayer data; familiarity with data produced by the IRS National Research Program (NRP) and ability to customize and/or apply the Rule Learner module source code to construct the demonstration module. The IRS Office of Research has identified OpenRules, Inc. of Edison, NJ as the sole source to provide the necessary services. Responsible sources who believe they can provide the required work may submit a written capability statement, via e-mail, no later than 15 days from the published date of this announcement. Telephone and fax inquiries will be considered non-responsive. No solicitation will be issued, therefore, requests for solicitations will be considered non-responsive. If no responses are received the award will be made to OpenRules, Inc.
- Web Link
-
FedBizOpps Complete View
(https://www.fbo.gov/?s=opportunity&mode=form&id=d3fca927caa944c8d1fc136e7ae0a62e&tab=core&_cview=1)
- Place of Performance
- Address: All work shall be performed at Contractor’s own facility except those government locations specifically identified in conjunction with one or more sub-tasks., United States
- Record
- SN01626576-W 20080801/080730221210-43f143749461d56e86b3eec3b8447aed (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 |