The Role of Text Analytics and Information Retrieval in the Accounting Domain

ABSTRACT: Textual documents proliferate throughout accounting practice, and a wide variety of groups make financial decisions based on written guidance. The Generally Accepted Accounting Principles (GAAP), along with annual corporate financial statements and other reporting narratives, are important sources of such guidance and information. This paper examines the literature in two major areas relevant to text analytics and information retrieval in the accounting domain: (1) the manual and computational content analysis of accounting narratives, accounting readability studies, and related text-mining work, and (2) the information retrieval literature stream that addresses the extraction of both text elements as well as quantities imbedded in text from accounting documents, and includes the impact of understanding the accounting lexicon upon retrieval from digital accounting documents. We use the goals in developing the GAAP Codification, as expressed by the Financial Accounting Standards Board (FASB) in their Notice to Constituents (FASB 2009), as a starting point for reviewing the literature. The paper concludes with a map for suggested future research in accounting text analytics and information retrieval.