Comparative Analysis of SVM Variants for GST Fraud Detection
Vivek Vyas · Communications on Applied Nonlinear Analysis · 2024
This study explores various Support Vector Machine (SVM) variants and provides an analytical comparison to identify their effectiveness in detecting Goods and Services Tax (GST) fraud. GST fraud poses significant challenges to regulatory authorities and businesses, necessitating robust detection methods. SVM, a powerful machine learning algorithm, offers promise in this domain due to its ability to handle complex data and nonlinear relationships. Through a comprehensive examination of SVM variants, including linear SVM, polynomial SVM, and radial basis function SVM, this study assesses their performance in GST fraud detection. Additionally, computational efficiency and scalability are investigated to gauge the practical viability of each variant. The findings contribute to advancing the understanding of SVM's applicability in fraud detection contexts and offer insights into selecting the most suitable variant for GST fraud identification. Ultimately, this research aids stakeholders, including tax authorities and businesses, in implementing effective strategies to combat fraudulent activities and uphold fiscal integrity.