Financial Statement Audit Using Support Vector Machines, Artificial Neural Networks and K-Nearest Neighbor: Empirical Study of UK and Ireland

Aram Khalaf Nawaiseh and Maysam F Abbod (Brunel University, United Kingdom (Great Britain)); Take Itagaki (Brunel University London, United Kingdom (Great Britain))

Initial applications of big data analytics such as data mining techniques in auditing remain pioneering, and more research is needed on attributes to develop predicted models using data mining analytics on financial statement auditing. This study explores data mining abilities based on Support Vector Machines (SVM), Artificial Neural Networks (ANN), and K-Nearest Neighbor (KNN) as predictive classification models for financial statement audit. Empirical results showed that ANN and SVM techniques achieved higher average accuracy rate and outperformed KNN in correctly classifying healthy companies. However, ANN had the lowest rate of Type I error, indicating better ability in classifying healthy companies' compeer to other techniques. SVM had better performance in terms of fewer incorrect classification of qualified companies into unqualified class, with the lowest rate of Type II error. This study demonstrates the superiority of ANN and SVM in predictive classification of correct auditing opinions. This is a pioneering auditing study using big data financial information to investigate attributes of developing prediction models.

Conference: UKSim-AMSS 22nd International Conference on Computer Modelling and Simulation, UKSim2020

Published: Mar 25, 2020

DOI: 10.5013/IJSSST.a.21.02.07