A skimming fraud detection model in automatic teller machine (ATM) transaction using machine learning : a case study at regional development bank in Indonesia = Model deteksi penipuan skimming pada transaksi anjungan tunai mandiri (ATM) menggunakan machine learning : studi kasus di bank pembangunan daerah di Indonesia

Pontororing, Natalie Christie (2023) A skimming fraud detection model in automatic teller machine (ATM) transaction using machine learning : a case study at regional development bank in Indonesia = Model deteksi penipuan skimming pada transaksi anjungan tunai mandiri (ATM) menggunakan machine learning : studi kasus di bank pembangunan daerah di Indonesia. Masters thesis, Universitas Pelita Harapan.

Preview	Text (Title) Title.pdf Available under License Creative Commons Attribution Non-commercial Share Alike. Download (437kB) | Preview
Preview	Text (Abstract) Abstract.pdf Available under License Creative Commons Attribution Non-commercial Share Alike. Download (444kB) | Preview
Preview	Text (ToC) ToC.pdf Available under License Creative Commons Attribution Non-commercial Share Alike. Download (440kB) | Preview
Preview	Text (Chapter1) Chapter1.pdf Available under License Creative Commons Attribution Non-commercial Share Alike. Download (410kB) | Preview
	Text (Chapter2) Chapter2.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (576kB)
	Text (Chapter3) Chapter3.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (793kB)
	Text (Chapter4) Chapter4.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (2MB)
	Text (Chapter5) Chapter5.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (411kB)
Preview	Text (Bibliography) Bibliography.pdf Available under License Creative Commons Attribution Non-commercial Share Alike. Download (414kB) | Preview
	Text (Appendices) Appendices.pdf Restricted to Repository staff only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (3MB)

Abstract

Fraud presents a significant challenge in the banking sector, leading to financial losses and eroding customer trust. Traditional fraud detection methods, such as manual auditing, are constrained by the auditors' expertise and the extensive amount of transaction data to be analyzed. To overcome these constraints, numerous organizations have adopted machine learning algorithms for fraud detection. This research delves into the application of Support Vector Machine (SVM), Random Forest, and Gradient Boosting algorithms to detect skimming fraud in ATM transactions at a Regional Development Bank in Indonesia. The high prevalence of fraudulent activities necessitates advanced techniques that can effectively pinpoint suspicious transactions and mitigate potential risks. Machine learning algorithms, such as SVM, Random Forest, and Gradient Boosting, have demonstrated impressive capabilities in detecting fraud by leveraging vast transaction data. The research findings suggest that Gradient Boosting outperforms the other models, with a recall score of 0.960, precision score of 0.966, F1 score of 0.963, and an AUC score of 0.998. Moreover, Gradient Boosting provides estimates of feature importance and adapts well to unseen data. The deployment of Gradient Boosting can bolster fraud detection capabilities, minimize financial losses, and enhance customer trust, particularly in the context of skimming fraud detection in ATM transactions at regional development banks in Indonesia. / Penipuan menimbulkan tantangan berarti dalam sektor perbankan, berdampak pada kerugian finansial dan merusak kepercayaan pelanggan. Batasan metode deteksi penipuan tradisional, seperti audit manual, terletak pada keahlian auditor serta jumlah data transaksi yang sangat besar untuk dianalisis. Untuk melampaui batasan ini, sejumlah organisasi telah memilih algoritma pembelajaran mesin sebagai alat deteksi penipuan. Penelitian ini memfokuskan diri pada penggunaan algoritma Support Vector Machine (SVM), Random Forest, dan Gradient Boosting dalam mendeteksi penipuan skimming dalam transaksi ATM di sebuah Bank Pembangunan Daerah di Indonesia. Tingginya prevalensi aktivitas penipuan membutuhkan teknik-teknik canggih yang dapat efektif dalam mengidentifikasi transaksi mencurigakan dan meredam risiko potensial. Algoritma-algoritma pembelajaran mesin seperti SVM, Random Forest, dan Gradient Boosting telah menunjukkan kinerja yang impresif dalam mendeteksi penipuan dengan memanfaatkan data transaksi dalam jumlah besar. Hasil penelitian menunjukkan keunggulan Gradient Boosting dibandingkan model lainnya, dengan skor recall 0.960, skor precision 0.966, skor F1 0.963, dan skor AUC 0.998. Lebih jauh, Gradient Boosting memberikan perkiraan pentingnya fitur dan dapat beradaptasi dengan baik pada data yang belum pernah dilihat. Penerapan Gradient Boosting dapat memperkuat kemampuan deteksi penipuan, meminimalisir kerugian finansial, dan meningkatkan kepercayaan pelanggan, khususnya dalam konteks deteksi penipuan skimming dalam transaksi ATM di bank pembangunan daerah di Indonesia.

Actions (login required)

View Item