Computer Expert Systems in Tax Practice

Date of Submission


Document Type


Degree Name

Master of Science in Accounting




Eleanor Fillebrown


Tax accounting--Data processing, Expert systems (Computer science)--United States

Call No. at the Univ. of New Haven Library

AS 36 .N29 Acc. 1992 no.7


Accountants have considerable difficulty collecting and interpreting information from clients, communicating precisely with one another, remaining informed, and applying recommended procedures when indicated. New methods of engineering knowledge bases, in general, and accounting knowledge bases, in particular, promise to revolutionize access to tax information and decision making in tax practice.

This thesis discusses how computers can help solve the information problems of tax practitioners and describes information management computer innovations. In particular, the paper discusses new expert systems for which direct evidence exists to show that they can help solve the information problems of tax practitioners. Several tax research techniques which are used currently by the tax practitioner are reviewed and their problems are discussed. State-of-the-art techniques being developed in the IRS Artificial Intelligence lab and the influence of these developments on future trends in the use of computers to facilitate taxation by government are discussed. Original contributions made by this thesis include: 1) writing of an executable computer program to demonstrate an expert system and 2) proposal of a new and efficient knowledge engineering method to construct and maintain a tax knowledge base.