Rewrites of an Expert System—Lessons Learned

Claude O. Archer and Peter Peer

Intelligent Systems in Accounting, Finance and Management, 1992, vol. 1, issue 2, 93-101

Abstract: One component of the 1986 Tax Law was to require taxpayers to file information returns on magnetic media if the quantity to be filed was sufficiently large. A waiver from this filing requirement was possible under hardship and other criteria, and could be obtained by filing a form with the Internal Revenue Service (IRS). Initially, the processing of waiver requests was accomplished manually by several relatively high‐level analytical personnel. The IRS's Artificial Intelligence Laboratory (AI Lab) created an Expert System to automate the process. An initial prototype was developed in OPS5+, an AI language, and the system was used on all waiver requests for Forms 1099 for Tax Year 1987, demonstrating feasibility and capability. The system was expanded in the following year, and used on waiver requests for all form types. For Tax Year 1989 the system was rewritten in LEVEL5, so that it could be more easily maintained by in‐house personnel. However, implementation problems forced a rapid rewrite, using CLIPPER, a dBase III+ compiler language. This version was subsequently used for all waiver requests for Tax Years 1989 and 1990. Presently, one clerical person can handle all of the waivers. This paper deals with the reasons and ramifications of this language‐shell‐language transition.

Date: 1992
References: Add references at CitEc
Citations:

Downloads: (external link)
https://doi.org/10.1002/j.1099-1174.1992.tb00011.x

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:wly:isacfm:v:1:y:1992:i:2:p:93-101

Ordering information: This journal article can be ordered from
http://www.blackwell ... bs.asp?ref=1099-1174

Access Statistics for this article

More articles in Intelligent Systems in Accounting, Finance and Management from John Wiley & Sons, Ltd.
Bibliographic data for series maintained by Wiley Content Delivery ().