Title: Financial fraud detection of the Egyptian companies annual reports using artificial bee colony algorithm
Authors: Essam Amin
Addresses: Faculty of Commerce, Damanhour University, Egypt
Abstract: Due to the increasing fraud cases, especially in the financial statements and its huge negative impact on the economy as a whole, along with the absence of clear codes among the Egyptian companies to detect fraud. This paper examines the use of artificial bee colony (ABC) algorithm for extracting the most suitable feature subset of accounting ratios to detect fraud on a sample of 83 Egyptian companies from 2010-2017 combined with several classifiers. Results show that the application of ABC has a significant effect on reducing the accounting ratios with higher accuracy along with decreasing incorrectly classified cases and reducing the required CPU computational time across six different classifiers.
Keywords: fraud detection; FD; artificial bee colony; ABC; decision tree; DT; Egypt; feature selection; FS.
DOI: 10.1504/IJBDA.2019.104163
International Journal of Business and Data Analytics, 2019 Vol.1 No.2, pp.184 - 201
Received: 05 Sep 2018
Accepted: 09 Feb 2019
Published online: 20 Dec 2019 *