Modeling the Tax Code

In this paper, we argue for the necessity of constructing an ontology which represents the U.S. Internal Revenue Code (IRC). Doing so would enable the construction of a broad range of intelligent applications, including automatic auditing software, robust on-line help systems, and tax question-answering systems. In this paper, we propose the construction of a rich ontology which models the tax code. We examine some of the unique challenges presented by a tax ontology and provide examples of the types of knowledge necessary for such an ontology.