Digital Taxation
Loading...
Date
2020
Authors
Wanjagi, Jedidah
Ruto, Kipkirui
Journal Title
Journal ISSN
Volume Title
Publisher
Kenya School of Revenue Administration
Abstract
This paper sought to determine how tax administrators could use data mining and pattern
recognition to enhance tax compliance on online business transactions. The specific objectives
of the study were to examine technology required in adopting data mining; to determine the tax
audit and control required to detect error and fraud in data mining; and to determine the risks
involved in data mining and pattern recognition to enhance tax compliance on online business
transactions. The latent role and benefits of data mining in tax administrations are elucidated
in view of the overall technology, operational framework and organization. The researcher
reviewed various articles, research papers and books on various data mining applications.
Techniques used for data mining included statistical techniques, decision tree and neuro
network technique. Findings indicate that decision tree and neural network technique provided
better results than the other techniques. The predictive modeling using the “Delphi” method
was discovered as perfect tool that assisted agency to differentiate non-compliance from
compliant clients and to focus on audits that would lead to a positive tax adjustment. The KRA
may consider the use of this model to predict the risk involved in data mining. This actually
assists the tax authorities to make better use of human personnel and therefore minimize the
tax burden. The process of data mining helps the tax administrators to refine its traditional audit
strategies in order to raise their tax budget
Description
ATCR Journal Article
Keywords
Data mining, Pattern recognition, Tax Compliance
Citation
Wanjagi, J., & Ruto, K. (2020). Digital Taxation. African Tax and Customs Review, 3(1), p30-38. Retrieved from https://atcr.kra.go.ke/ojs/index.php/atcr/article/view/63