{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,13]],"date-time":"2026-06-13T16:38:37Z","timestamp":1781368717189,"version":"3.54.1"},"reference-count":40,"publisher":"MDPI AG","issue":"6","license":[{"start":{"date-parts":[[2022,5,31]],"date-time":"2022-05-31T00:00:00Z","timestamp":1653955200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Future Internet"],"abstract":"<jats:p>Detecting tax fraud is a top objective for practically all tax agencies in order to maximize revenues and maintain a high level of compliance. Data mining, machine learning, and other approaches such as traditional random auditing have been used in many studies to deal with tax fraud. The goal of this study is to use Artificial Neural Networks to identify factors of tax fraud in income tax data. The results show that Artificial Neural Networks perform well in identifying tax fraud with an accuracy of 92%, a precision of 85%, a recall score of 99%, and an AUC-ROC of 95%. All businesses, either cross-border or domestic, the period of the business, small businesses, and corporate businesses, are among the factors identified by the model to be more relevant to income tax fraud detection. This study is consistent with the previous closely related work in terms of features related to tax fraud where it covered all tax types together using different machine learning models. To the best of our knowledge, this study is the first to use Artificial Neural Networks to detect income tax fraud in Rwanda by comparing different parameters such as layers, batch size, and epochs and choosing the optimal ones that give better accuracy than others. For this study, a simple model with no hidden layers, softsign activation function performs better. The evidence from this study will help auditors in understanding the factors that contribute to income tax fraud which will reduce the audit time and cost, as well as recover money foregone in income tax fraud.<\/jats:p>","DOI":"10.3390\/fi14060168","type":"journal-article","created":{"date-parts":[[2022,5,31]],"date-time":"2022-05-31T05:25:42Z","timestamp":1653974742000},"page":"168","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":30,"title":["Fraud Detection Using Neural Networks: A Case Study of Income Tax"],"prefix":"10.3390","volume":"14","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-7798-1781","authenticated-orcid":false,"given":"Belle Fille","family":"Murorunkwere","sequence":"first","affiliation":[{"name":"African Center of Excellence in Data Science, University of Rwanda, KK 737 Street, Gikondo, Kigali P.O. Box 4285, Rwanda"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-2089-0939","authenticated-orcid":false,"given":"Origene","family":"Tuyishimire","sequence":"additional","affiliation":[{"name":"African Institute for Mathematical Sciences, KN 3 Street, Remera, Kigali P.O. Box 7150, Rwanda"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Dominique","family":"Haughton","sequence":"additional","affiliation":[{"name":"Department of Mathematical Sciences and Global Studies, Bentley University, Watham, MA 02452-4705, USA"},{"name":"Department of Mathematical Sciences and Global Studies, Universit\u00e9 Paris 1 (SAMM), 75634 Paris, France"},{"name":"Department of Mathematical Sciences and Global Studies, Universit\u00e9 Toulouse 1 (TSE-R), 31042 Toulouse, France"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-2973-9258","authenticated-orcid":false,"given":"Joseph","family":"Nzabanita","sequence":"additional","affiliation":[{"name":"Department of Mathematics, College of Science and Technology, University of Rwanda, KN 67 Street, Nyarugenge, Kigali P.O. Box 3900, Rwanda"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1968","published-online":{"date-parts":[[2022,5,31]]},"reference":[{"key":"ref_1","unstructured":"Smelser, N.J., and Baltes, P.B. (2011). International Encyclopedia of the Social & Behavioral Sciences, Elsevier. [11th ed.]."},{"key":"ref_2","unstructured":"de la Feria, R. (2018). Tax Fraud and the Rule of Law, Oxford University Centre for Business Taxation."},{"key":"ref_3","unstructured":"(2021, August 25). Tax Evasion Most Prevalent Financial Crime in Rwanda. Available online: https:\/\/www.newtimes.co.rw\/news\/tax-evasion-most-prevalent-financial-crime-rwanda."},{"key":"ref_4","unstructured":"(2021, August 16). Using Analytics Successfully to Detect Fraud. Available online: https:\/\/assets.kpmg\/content\/dam\/kpmg\/pdf\/2016\/07\/using-analytics-sucessfully-to-detect-fraud.pdf."},{"key":"ref_5","first-page":"12191","article-title":"Tax fraud and the rule of law","volume":"34","year":"2017","journal-title":"Expert Syst."},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"1427","DOI":"10.1016\/j.eswa.2012.08.051","article-title":"Characterization and detection of taxpayers with false invoices using data mining techniques","volume":"40","year":"2013","journal-title":"Expert Syst. Appl."},{"key":"ref_7","unstructured":"Dias, A., Pinto, C., Batista, J., and Neves, E. (2016). Signaling tax evasion, financial ratios and cluster analysis. BIS Q. Rev."},{"key":"ref_8","doi-asserted-by":"crossref","first-page":"8769","DOI":"10.1016\/j.eswa.2012.01.204","article-title":"Using data mining technique to enhance tax evasion detection performance","volume":"10","author":"Wu","year":"2012","journal-title":"Expert Syst. Appl."},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"35","DOI":"10.1016\/j.gltp.2021.01.006","article-title":"Credit card fraud detection using Artificial Neural Networks","volume":"2","author":"Asha","year":"2021","journal-title":"Glob. Transitions Proc."},{"key":"ref_10","doi-asserted-by":"crossref","unstructured":"Ghosh, S., and Douglas, L.R. (1994, January 4\u20137). Credit card fraud detection with a neural-network. Proceedings of the Twenty-Seventh Hawaii International Conference, Wailea, HI, USA.","DOI":"10.1109\/HICSS.1994.323314"},{"key":"ref_11","doi-asserted-by":"crossref","unstructured":"Mubarek, A.M., and E\u015fref, A. (2017, January 5\u20138). CMultilayer perceptron neural network technique for fraud detection. Proceedings of the S2017 International Conference on Computer Science and Engineering (UBMK), Antalya, Turkey.","DOI":"10.1109\/UBMK.2017.8093417"},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"291","DOI":"10.1023\/A:1009700419189","article-title":"Adaptive fraud detection","volume":"1","author":"Fawcett","year":"1997","journal-title":"Data Min. Knowl. Discov."},{"key":"ref_13","doi-asserted-by":"crossref","unstructured":"Bonchi, F., Giannotti, F., Mainetto, G., and Pedreschi, D. (1999, January 30). Using data mining techniques in fiscal fraud detection. In Proceedings of the International Conference on Data Warehousing and Knowledge Discovery, Berlin\/Heidelberg, Germany.","DOI":"10.1007\/3-540-48298-9_39"},{"key":"ref_14","doi-asserted-by":"crossref","unstructured":"de Roux, D., Perez, B., Moreno, A., Villamil, M.D.P., and Figueroa, C. (2018, January 19\u201323). Tax fraud detection for under-reporting declarations using an unsupervised machine learning approach. Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, London, UK.","DOI":"10.1145\/3219819.3219878"},{"key":"ref_15","doi-asserted-by":"crossref","unstructured":"P\u00e9rez L\u00f3pez, C., Delgado Rodr\u00edguez, M., and de Lucas Santos, S. (2019). Tax fraud detection through neural networks: An application using a sample of personal income taxpayers. Future Internet, 11.","DOI":"10.3390\/fi11040086"},{"key":"ref_16","doi-asserted-by":"crossref","unstructured":"Savi\u0107, M., Atanasijevi\u0107, J., Jakoveti\u0107, D., and Kreji\u0107, N. (2021). Tax Evasion Risk Management Using a Hybrid Unsupervised Outlier Detection Method. arXiv.","DOI":"10.1016\/j.eswa.2021.116409"},{"key":"ref_17","doi-asserted-by":"crossref","unstructured":"Neagoe, V.-E., Ciotec, A.-D., and Cucu, G.-S. (2018, January 14\u201316). Deep convolutional neural networks versus multilayer perceptron for financial prediction. Proceedings of the 2018 International Conference on Communications (COMM), Bucharest, Romania.","DOI":"10.1109\/ICComm.2018.8453730"},{"key":"ref_18","doi-asserted-by":"crossref","unstructured":"McCulloch, W.S., and Pitts, W. (1943). A Logical Calculus of the Ideas Immanent in Nervous Activity, Springer.","DOI":"10.1007\/BF02478259"},{"key":"ref_19","unstructured":"G\u00e9ron, A. (2019). Hands-on Machine Learning with Scikit-Learn, Keras, and TensorFlow: Concepts, Tools, and Techniques to Build Intelligent Systems, O\u2019Reilly Media."},{"key":"ref_20","unstructured":"Abraham, A. (2005). Artificial Neural Networks, John Wiley & Sons, Ltd."},{"key":"ref_21","unstructured":"(2020, July 31). Math behind Artificial Neural Networks. Available online: https:\/\/medium.com\/analytics-vidhya\/math-behind-artificial-neural-networks-42f260fc1b25."},{"key":"ref_22","unstructured":"Mohamed, H., Negm, A., Zahran, M., and Saavedra, O.C. (2015, January 12\u201314). Assessment of Artificial Neural Networks for Bathymetry Estimation Using High Resolution Satellite Imagery in Shallow Lakes: Case Study El Burullus Lake. Proceedings of the Eighteenth International Water Technology Conference, IWTC18 Sharm, ElSheikh, Egypt."},{"key":"ref_23","doi-asserted-by":"crossref","unstructured":"Sharma, S., Sharma, S., and Athaiya, A. (2017). Activation Functions in Neural Networks, Towards Data Science. Available online: http:\/\/ijeast.com\/papers\/310-316,Tesma412,IJEAST.pdf.","DOI":"10.33564\/IJEAST.2020.v04i12.054"},{"key":"ref_24","unstructured":"Chollet, F. (2021). Deep Learning with Python, Simon and Schuster."},{"key":"ref_25","unstructured":"Agostinelli, F., Hoffman, M., Sadowski, P., and Baldi, P. (2014). Learning Activation Functions to Improve Deep Neural Networks. arXiv."},{"key":"ref_26","unstructured":"Dangeti, P. (2017). Statistics for Machine Learning, Packt Publishing Ltd."},{"key":"ref_27","doi-asserted-by":"crossref","unstructured":"Lin, G., and Shen, W. (2018). Research on Convolutional Neural Network Based on Improved Relu Piecewise Activation Function, Elsevier.","DOI":"10.1016\/j.procs.2018.04.239"},{"key":"ref_28","unstructured":"Anthadupula, S.P., and Gyanchandani, M. (2021). A Review and Performance Analysis of Non-Linear Activation Functions in Deep Neural Networks. Int. Res. J. Mod. Eng. Technol. Sci."},{"key":"ref_29","doi-asserted-by":"crossref","unstructured":"Zheng, H., Yang, Z., Liu, W., Liang, J., and Li, Y. (2015). Improving Deep Neural Networks Using Softplus Units, IEEE.","DOI":"10.1109\/ChinaSIP.2014.6889194"},{"key":"ref_30","unstructured":"(2018, July 13). Difference between a Batch and an Epoch in a Neural Network. Available online: https:\/\/machinelearningmastery.com\/difference-between-a-batch-and-an-epoch\/."},{"key":"ref_31","doi-asserted-by":"crossref","unstructured":"Goutte, C., and Gaussier, E. (2005). A Probabilistic Interpretation of Precision, Recall and F-Score, With Implication for Evaluation, Springer.","DOI":"10.1007\/978-3-540-31865-1_25"},{"key":"ref_32","doi-asserted-by":"crossref","first-page":"5052","DOI":"10.1214\/17-EJS1338SI","article-title":"Beyond Sigmoids: How to Obtain Well-Calibrated Probabilities from Binary Classifiers with Beta Calibration","volume":"11","author":"Kull","year":"2017","journal-title":"Electron. J. Stat."},{"key":"ref_33","unstructured":"Kingma, D.P., and Ba, J. (2014). Adam: A Method for Stochastic Optimization. arXiv."},{"key":"ref_34","doi-asserted-by":"crossref","unstructured":"Heaton, J., McElwee, S., Fraley, J., and Cannady, J. (2017). Early Stabilizing Feature Importance for TensorFlow Deep Neural Networks, IEEE.","DOI":"10.1109\/IJCNN.2017.7966442"},{"key":"ref_35","doi-asserted-by":"crossref","unstructured":"de S\u00e1, C.R. (2019, January 28\u201330). Variance-based feature importance in neural networks. Proceedings of the 22nd International Conference, DS 2019, Split, Croatia.","DOI":"10.1007\/978-3-030-33778-0_24"},{"key":"ref_36","doi-asserted-by":"crossref","unstructured":"Zhou, Z., Zheng, W.-S., Hu, J.-F., Xu, Y., and You, J. (2016). One-Pass Online Learning: A Local Approach, Elsevier.","DOI":"10.1016\/j.patcog.2015.09.003"},{"key":"ref_37","unstructured":"Garavaglia, S., and Sharma, A. (1998, January 4\u20136). A smart guide to dummy variables: Four applications and a macro. Proceedings of the Northeast SAS Users Group Conference, Pittsburgh, PA, USA."},{"key":"ref_38","doi-asserted-by":"crossref","unstructured":"Kaur, P., and Gosain, A. (2018). Comparing the behavior of oversampling and undersampling approach of class imbalance learning by combining class imbalance problem with noise. ICT Based Innovations, Springer.","DOI":"10.1007\/978-981-10-6602-3_3"},{"key":"ref_39","doi-asserted-by":"crossref","first-page":"321","DOI":"10.1613\/jair.953","article-title":"SMOTE: Synthetic minority over-sampling technique","volume":"16","author":"Chawla","year":"2002","journal-title":"J. Artif. Intell. Res."},{"key":"ref_40","doi-asserted-by":"crossref","unstructured":"Murorunkwere, B.F., Dominique, H., Nzabanita, J., and Kipkogei, F. (2022). Predicting Tax Fraud Using Supervised Machine Learning Approach. Afr. J. Sci. Technol. Innov. Dev., submitted.","DOI":"10.1080\/20421338.2023.2187930"}],"container-title":["Future Internet"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/1999-5903\/14\/6\/168\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,10]],"date-time":"2025-10-10T23:22:29Z","timestamp":1760138549000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/1999-5903\/14\/6\/168"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2022,5,31]]},"references-count":40,"journal-issue":{"issue":"6","published-online":{"date-parts":[[2022,6]]}},"alternative-id":["fi14060168"],"URL":"https:\/\/doi.org\/10.3390\/fi14060168","relation":{},"ISSN":["1999-5903"],"issn-type":[{"value":"1999-5903","type":"electronic"}],"subject":[],"published":{"date-parts":[[2022,5,31]]}}}