Temporal reconstruction of financial accounting standards: A prototype system for pension accounting

The accounting knowledge base continues to expand at an ever-increasing rate. The increase in the number of rules that comprise the body of accounting standards is accompanied by an increase in both the complexity of the rules and an increase in the fragmentation of their application. The task of effectively and efficiently navigating this knowledge base is a daunting task. Software methodologies are proposed as a means to improve the understandability of standards and facilitating access to standards. This thesis offers a structure for a prototype financial accounting standards retrieval system to support the temporal reconstruction of a specific subset of the financial accounting standards, the pension standards, part of the accounting standards literature more commonly referred to as Generally Accepted Accounting Principles (GAAP). The structure and design are implemented as a stand-alone prototype application that automatically temporally reconstructs a user-specified standard at a user-specified date.; Accounting standards are revised and rescinded continuously. Such changes occur either as part of a newly promulgated standard or result in the issuance of a new standard whose sole purpose is to detail the amendment. The result is a multi-layered knowledge base. Professional accountants are left with a complex manual task of re-building standards amended over time. This thesis introduces a knowledge representation schema comprised of a taxonomy for understanding and organizing change as it occurs in financial accounting standards. The prototype offers a conceptual UML model for financial accounting standards with implications for standards engineering and knowledge base design. XML Document Type Definitions provide standards description, mark-up and support the identification and retrieval of accounting standard changes. The programmed reconstruction procedure permits automatic reconstruction of a financial accounting standard at any user specified date, thereby affording temporal access to financial accounting standards. The thesis contributes to the ontological representation of change in the accounting domain, to the representation of changes in complex objects in the knowledge representation domain and to the development of systems using meta-language representations in artificial intelligence. Recommendations for the restructuring of financial accounting standards as well as for the future development of refined XML standards for accounting standards are offered.