Application of Feature Engineering with Classification Techniques to Enhance Corporate Tax Default Detection Performance
The objective of this work is to propose a methodology that is helpful in analyzing tax data and predict significant features that cause tax defaulting. In this work, we gathered a Finnish tax default data of different firms and then split it according to
- PDF / 369,960 Bytes
- 11 Pages / 439.37 x 666.142 pts Page_size
- 47 Downloads / 200 Views
Md. S. Satu Department of MIS, Noakhali Science and Technology University, Noakhali, Bangladesh M. Zoynul Abedin (B) Department of Finance and Banking, Hajee Mohammad Danesh Science and Technology University, Dinajpur 5200, Bangladesh e-mail: [email protected] S. Khanom Department of Electronics and Communication Engineering, Institute of Science Trade and Technology, Dhaka, Bangladesh J. Ouenniche University of Edinburgh, Business School, 29 Buccleuch Place, Edinburgh EH8 9JS, UK M. Shamim Kaiser Institute of Information Technology, Jahangirnagar University, Dhaka, Bangladesh © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2021 M. S. Kaiser et al. (eds.), Proceedings of International Conference on Trends in Computational and Cognitive Engineering, Advances in Intelligent Systems and Computing 1309, https://doi.org/10.1007/978-981-33-4673-4_5
53
54
S. Satu et al.
1 Introduction Tax is a primary source of government earning that allocates funds to the various social plans, infrastructure and services. According to the World Bank studies, the average universal paid taxes are almost 40% where taxes of the default firms might not be recovered in the forthcoming years. But, only a small amount of works were happened in view of unpaid tax statistics. Now-a-days every country gives attention to obtain tax revenue because of planning budget [7, 10]. In Finland, 12% active firms were not paid any taxes at the end of 2015 [6]. However, tax audits were assessed the risk of different firms using the likelihood of tax default. Alongside, inability of paying tax may decline the economical state that causes raising of workless people, pretentious micro and macroeconomic consequences, communal displacement, monetary depressions and collapses. Recently, machine learning is an emergent field that resolves various forecasting issues in the finance and accounting sector. It produces laborious, potentially novel and comprehensive findings to make decision. Therefore, the current work is focused on machine learning based modeling that identify tax default at different firms. There were happened a small amount of works in tax default. In the previous studies [6] they employed machine learning to analyze tax default conditions and focused only single feature selection and classification methods such as genetic algorithm (GA) and linear discriminant analysis (LDA) respectively. However, we extended this work with more machine learning methods and explored significant features about tax default. In the empirical point of view, machine learning was used massively to explore this issue that was not happened in previous studies. Again, feature transformation and selection methods were provided more appropriate tax default prediction. Moreover, non parametric statistical test is re-evaluated the findings about this work. In the managerial perspective, it is useful to reduce their work for the tax authorities and stakeholders. This automatic solution can forecast which organization will be failed to give taxes
Data Loading...
