Foundations and Trends® in Accounting > Vol 17 > Issue 2

Bookkeeping Graphs: Computational Theory and Applications

By Pierre Jinghong Liang, Carnegie Mellon University, USA and University of Hong Kong, Hong Kong, liangj@andrew.cmu.edu

 
Suggested Citation
Pierre Jinghong Liang (2023), "Bookkeeping Graphs: Computational Theory and Applications", Foundations and Trends® in Accounting: Vol. 17: No. 2, pp 77-172. http://dx.doi.org/10.1561/1400000070

Publication Date: 18 Apr 2023
© 2023 P. J. Liang
 
Subjects
Auditing,  Financial statement analysis and equity valuation,  Corporate governance,  Management control,  Data compression,  Pattern recognition and learning,  Detection and estimation,  Summarization,  Data mining
 

Free Preview:

Download extract

Share

Download article
In this article:
1. Introduction and Overview
2. Bookkeeping Graphs and MDL
3. Pattern Recognition in Bookkeeping Data
4. Summary and Future Work
Acknowledgments
Appendices
References

Abstract

This monograph first describes the graph or network representation of Double-Entry bookkeeping both in theory and in practice. The representation serves as the intellectual basis for a series of applied computational works on pattern recognition and anomaly detection in corporate journal-entry audit settings. The second part of the monograph reviews the computational theory of pattern recognition and anomaly detection built on the Minimum Description Length (MDL) principle. The main part of the monograph describes how the computational MDL theory is applied to recognize patterns and detect anomalous transactions in graphs representing the journal entries of a large set of transactions extracted from real-world corporate entities’ bookkeeping data.

DOI:10.1561/1400000070
ISBN: 978-1-63828-164-1
112 pp. $80.00
Buy book (pb)
 
ISBN: 978-1-63828-165-8
112 pp. $150.00
Buy E-book (.pdf)
Table of contents:
1. Introduction and Overview
2. Bookkeeping Graphs and MDL
3. Pattern Recognition in Bookkeeping Data
4. Summary and Future Work
Acknowledgments
Appendices
References

Bookkeeping Graphs: Computational Theory and Applications

Bookkeeping Graphs: Computational Theory and Applications first describes the graph or network representation of Double-Entry bookkeeping both in theory and in practice. The representation serves as the intellectual basis for a series of applied computational works on pattern recognition and anomaly detection in corporate journal-entry audit settings. The second part of the monograph reviews the computational theory of pattern recognition and anomaly detection built on the Minimum Description Length (MDL) principle. The main part of the monograph describes how the computational MDL theory is applied to recognize patterns and detect anomalous transactions in graphs representing the journal entries of a large set of transactions extracted from real-world corporate entities’ bookkeeping data.

 
ACC-070