Clustering Model of False Positive Elimination in Moroccan Fiscal Fraud Detection
- 26 Downloads
Every year government faces tax fraud with many methods. Big data and analytics enable government organizations to improve existing processes of detection and engage in entirely new types of analyses to allow tax authority to prevent tax fraud, reduce the cost of managing taxes and optimize public spending. Fraud detection often includes analyzing large datasets to locate irregularities. Anomaly detection helps by exploring every possible path to find fraudsters but it’s often results in large number of false positive, that is entries wrongly identified as fraud. When a taxpayer is wrongly flagged, the intervention of an audit agent is necessary and so a big time of investigations to finally lose time, effort and money.
The purpose of this paper is to propose a model of false positive elimination in Moroccan fiscal fraud detection based on clustering.
KeywordsAnomaly detection False positive elimination Tax fraud
- 3.Hu, L., Li, T., Xie, N., Hu, J.: False positive elimination in intrusion detection based on clustering. In: 2015 12th International Conference on Fuzzy Systems and Knowledge Discovery, FSKD 2015, pp. 519–523 (2016). https://doi.org/10.1109/FSKD.2015.7381996
- 4.Kotu, V., Deshpande, B.: Anomaly detection. Data Sci. 447–6510 (2019). https://doi.org/10.1016/B978-0-12-814761-0.00013-7
- 7.Matos, T., de Macedo, J.A., Monteiro, J.M.: An empirical method for discovering tax fraudsters: a real case study of Brazilian fiscal evasion (2015). https://doi.org/10.1145/2790755.2790759
- 12.Chandola, V., Banerjee, A., Kumar, V.: Anomaly detection: a survey. Conform. Predict. Reliab. Mach. Learn.: Theory Adapt. Appl. 41(3), 71–97 (2014). https://doi.org/10.1016/B978-0-12-398537-8.00004-3CrossRefGoogle Scholar
- 14.Yao, Y., Yang, W., Gao F., Yu, G.: Anomaly intrusion detection approach using hybrid MLP/CNN neural. In: Network Proceedings of the Sixth International Conference on Intelligent Systems Design and Applications (ISDA 2006), 0-7695-2528-8/06 (2006)Google Scholar