Pengembangan Sistem Informasi Akuntansi Berbasis Machine Learning untuk Deteksi Dini Penghindaran Pajak pada Perusahaan Terbuka Indonesia

Abstract

This research focuses on developing a machine learning-based artificial intelligence architecture to serve as an automated detection instrument for potential tax avoidance among Indonesian publicly traded companies. The exploitation of tax regulation loopholes remains a critical challenge that massively degrades national revenue. The study population comprises corporate entities listed on the Indonesia Stock Exchange (IDX), selected via a purposive sampling approach. Information gathering utilized a dual-method strategy. Secondary data, extracted from historical financial databases within the LSEG Refinitiv Workspace, incorporating parameters such as the Effective Tax Rate (ETR), Cash Effective Tax Rate (CETR), and Debt to Equity Ratio (DER), served as the training dataset for the AI algorithm. Concurrently, primary data was obtained through in-depth interviews with four tax professionals and one academic to qualitatively map out supervisory constraints. The primary methodology involved formulating an anomaly detection model embedded within a web-based prototype to enable real-time assessments. The system's robustness underwent rigorous evaluation through a series of technical trials, encompassing reliability, compliance, recovery, and stress testing. Analytical outcomes demonstrate the model's high proficiency in precisely identifying concealed financial anomalies and suspicious patterns. The implementation of artificial intelligence not only elevates detection accuracy but also drastically accelerates conventional monitoring processes, thereby empowering fiscal authorities to adopt data-driven decision-making practices. Ultimately, the integration of machine learning presents a strategic solution to mitigate tax revenue leakage and optimize state income, while also paving the way for future research explorations concerning the adoption of cutting-edge technology in financial sector governance and tax administration.