RANCANG BANGUN ARTIFICIAL INTELLIGENCE DALAM DETEKSI OTOMATIS INDIKASI PENGHINDARAN PAJAK OLEH PERUSAHAAN: KASUS PERUSAHAAN TERBUKA DI INDONESIA

Rizal, Moch Rifqi (2025) RANCANG BANGUN ARTIFICIAL INTELLIGENCE DALAM DETEKSI OTOMATIS INDIKASI PENGHINDARAN PAJAK OLEH PERUSAHAAN: KASUS PERUSAHAAN TERBUKA DI INDONESIA. Skripsi thesis, Politeknik Keuangan Negara STAN.

	Text (Cover) 01. Cover_MOCH RIFQI RIZAL_4132230047.pdf Restricted to Repository staff only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (147kB)
	Text (Abstrak) 02. Abstrak_MOCH RIFQI RIZAL_4132230047.pdf Available under License Creative Commons Attribution Non-commercial Share Alike. Download (162kB)
	Text (Daftar Isi) 03. Daftar Isi_MOCH RIFQI RIZAL_4132230047.pdf Restricted to Repository staff only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (380kB)
	Text (Bab I) 05. Bab I_MOCH RIFQI RIZAL_4132230047.pdf Restricted to Repository staff only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (290kB)
	Text (Bab II) 06. Bab II_MOCH RIFQI RIZAL_4132230047.pdf Restricted to Repository staff only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (1MB)
	Text (Bab III) 07. Bab III_MOCH RIFQI RIZAL_4132230047.pdf Restricted to Repository staff only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (166kB)
	Text (Bab IV) 08. Bab IV_MOCH RIFQI RIZAL_4132230047.pdf Restricted to Repository staff only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (2MB)
	Text (Daftar Pustaka) 10. Daftar Pustaka_MOCH RIFQI RIZAL_4132230047.pdf Restricted to Repository staff only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (454kB)
	Text (Bab V) 09. Bab V_MOCH RIFQI RIZAL_4132230047.pdf Restricted to Repository staff only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (145kB)

Abstract

ABSTRAK Penelitian ini merancang model kecerdasan buatan berbasis machine learning untuk secara otomatis mendeteksi indikasi penghindaran pajak pada perusahaan terbuka di Indonesia. Penghindaran pajak menjadi isu strategis yang berpengaruh besar terhadap penerimaan negara karena celah hukum kerap dimanfaatkan untuk mengurangi beban paja k. Model deteksi anomali dikembangkan melalui integrasi algoritma machine learning dan diimplementasikan pada prototipe aplikasi web untuk analisis real time. Kinerja sistem diuji dengan reliability, compliance, recovery, dan stress testing, menunjukkan keandalan model dalam mengidentifikasi pola transaksi mencurigakan dalam laporan keuangan perusahaan. Hasil penelitian mengungkap bahwa pendekatan AI ini mampu meningkatkan akurasi pendeteksian, mempercepat proses pengawasan, serta mendukung pengambilan keputusan berbasis data bagi otoritas pajak. Selain berkontribusi pada peningkatan kepatuhan dan optimalisasi penerimaan pajak, temuan ini juga membuka peluang riset lanjutan dalam penerapan teknologi canggih di sektor keuangan dan perpajakan. Kata kunci: penghindaran pajak, kecerdasan buatan, machine learning, deteksi anomali, perpajakan A BSTRACT This study designs an artificial intelligence model based on machine learning to automatically detect indications of tax avoidance in publicly listed companies in Indonesia. Tax avoidance is a strategic issue that significantly impacts government revenue due to legal loopholes often used to reduce tax burdens. An anomaly detection model was developed by integrating machine learning algorithms and implemented in a web- based prototype for real- time analysis. System performance was validated through reliability, compliance, recovery, and stress testing, demonstrating the model’s robustness in identifying suspicious transaction patterns in corporate financial reports. Findings reveal that this AI -driven approach significantly enhances detection accuracy, acceler ates monitoring processes, and supports data-driven decision -making for tax authorities. Beyond boosting tax compliance and revenue optimization, the study also highlights new research prospects for advanced technology adoption in the finance and taxation sectors. Keywords: tax avoidance, artificial intelligence, machine learning, anomaly detection, taxation

Actions (login required)

View Item