Fraud Detection using Machine Learning and Deep Learning

Raghavan, P.; Gayar, Neamat El

doi:10.1109/iccike47802.2019.9004231

Cited by 94 publications

(40 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Mengingat maraknya penipuan pada situs e-commerce yang dapat mengakibatkan kerugian finansial yang cukup besar, sebagai konsumen perlu adanya pengetahuan mengenai jenis penipuan yang umum terjadi dan metode pencegahan yang digunakan untuk mendeteksi penipuan agar terhindar dari berbagai kerugian. Beberapa penelitian sebelumnya hanya membahas tentang identifikasi dan metode pencegahan penipuan e-commerce ( Makarti, 2011;Chang & Chang, 2012;Syed & Shabbir, 2013;Valentin, 2013;Caldeira, Brandao, & Pereira, 2014;Leung, Lai, Chen, & Wan, 2014;Massa & Valverde, 2014;Hwang & Lai, 2015;JRana & Baria, 2015;Singh & Singh, 2015;Abdallah, Maarof, & Zainal, 2016;Beránek, Nýdl, & Remeš, 2016;Gerlach, Pavlovic, & Gerlach, 2016;Lima & Pereira, 2016;Yang et al, 2016;Ramadhan & Amelia, 2016;Sun et al, 2017;Prisha, Neo, Ong, & Teo, 2017;Raghava-Raju, 2017;Shaji & Panchal, 2017;Wiralestari, 2017;Renjith, 2018;Weng et al, 2018;Zhao et al, 2018;Zheng et al, 2018); Amasiatu Amiruddin et al, 2019;Carta et al, 2019;Raghavan & Gayar, 2019;Shah et al, 2019;Soomro et al, 2019. Sementara penelitian lainnya lebih fokus pada penipuan sistem pembayaran dan penipuan terkait dengan pelanggan (Keraf & Hidup, 2010;Rofiq & Mula, 2010;Raj & Portia, 2011;Hu, Liu, & Sambamurthy, 2011;…”

Section: Pendahuluanunclassified

Pemetaan Bentuk Dan Pencegahan Penipuan E-Commerce

Dewi

Mahyuni

2020

EEB

View full text Add to dashboard Cite

Seiring berkembangnya transaksi jual beli, penipuan elektronik juga turut meningkat sehingga mengakibatkan banyak konsumen yang telah mengalami kerugian akibat penipuan yang terjadi. Tujuan dari penelitian ini adalah untuk mengetahui jenis penipuan yang dapat terjadi dalam perdagangan elektronik dan pencegahan yang dapat dilakukan. Dalam penelitian ini disajikan review dengan metode pendekatan interpretif atas artikel terkait, dengan proses pemetaan pada artikel yang dikumpulkan melalui situs Google Cendekia, Elsevier, Springer, Taylor & Francis, dan MDPI (Multidisciplinary Digital Publishing Institute). Dari sumber tersebut, 105 artikel berhasil dikumpulkan, setelah proses seleksi artikel berdasarkan 10 tahun terahir dan kesesuaian pembahasan akhirnya diperoleh 55 artikel. Hasil penelitian ini adalah ditemukan berbagai jenis penipuan pada keempat kategori e-coomerce serta penipuan pada sistem pebayaran dan penipuan pada e-commerce yang menyangkut pelanggan. Metode modern pendeteksi penipuan juga disajikan dalam penelitian ini, seperti data mining, jaringan bayesan, algoritma, mesin pendukung vector, pemrograman genetik, pohon pengambilan keputusan, Adaptive Neuro-Fuzzy Inference System, Situs Web Bantuan Perlindungan (PAW), dan Model Privacy Antecedent-Privacy Concern-Outcomes (APCO). Dengan penjabaran pada hasil penelitian ini, konsumen diharapkan untuk lebih berhati-hati saat melakukan transaksi di situs e-commerce agar terhindar dari berbagai tindak penipuan.

show abstract

Section: Pendahuluanunclassified

Pemetaan Bentuk Dan Pencegahan Penipuan E-Commerce

Dewi

Mahyuni

2020

EEB

View full text Add to dashboard Cite

show abstract

“…Vidanelage et al (2019) discussed varios machine learning techniques using "Scikit-learn Package in Python" to find the fraudulent transactions in dataset related payment [20]. Raghavan et al (2019) use the European (EU) Australian and German dataset and aim to benchmark deep learning and supervised machine learning methods [21]. Seemakurthi et al (2015) describe a new approach using texts classifers to detect fraudulent texts on text-based financial documents [22].…”

Section: Related Workmentioning

confidence: 99%

Karşılaştırmalı Veri Madenciliği Teknikleri Kullanılarak Bir E-Ticaret Sitesinin Kredi Kartı Dolandırıcılığı Sınıflandırması

Kırelli¹,

Arslankaya²,

Zeren³

2020

European Journal of Science and Technology

View full text Add to dashboard Cite

Credit card payment is one of the most preferred methods of e-commerce sites. Fraud orders are the biggest concerns for online shopping sites. Fraud operations affect not only customers but also companies and banks. Hence, companies should be able to classify orders and take measures against suspicious transactions. Classification is easier on the banking side because of more information about customers, but it is more difficult to determine this process on e-commerce sites. In this study, the actual order data of a private e-commerce enterprise has been examined and suspicious transactions are determined. First of all, all order data is analyzed and filtered. The best variables for classification are determined by variable selection algorithms. Afterwards, classification algorithms are applied and suspicious orders are determined with 92% success rate. Naïve Bayesian, Decision Trees and Artificial Neural Network have been used as comparative data mining methods.

show abstract

“…However, these efficiencies are often accompanied by increased complexity, and thus, raising concerns about the reliability of modern networks. Excessive use of these black-box predictors for critical topics such as medical diagnosis [3], [4], and particularly this case (fraud detection) [5], [6], [7], has made it extremely necessary to draw a line between well-grounded predictions and those with challenging inference. The lack of interpretability of the mentioned complex models is by no means the only issue.…”

Section: Introductionmentioning

confidence: 99%

Uncertainty-Aware Credit Card Fraud Detection Using Deep Learning

Habibpour¹,

Gharoun²,

Mehdipour³

et al. 2021

Preprint

View full text Add to dashboard Cite

Countless research works of deep neural networks (DNNs) in the task of credit card fraud detection have focused on improving the accuracy of point predictions and mitigating unwanted biases by building different network architectures or learning models. Quantifying uncertainty accompanied by point estimation is essential because it mitigates model unfairness and permits practitioners to develop trustworthy systems which abstain from suboptimal decisions due to low confidence. Explicitly, assessing uncertainties associated with DNNs predictions is critical in real-world card fraud detection settings for characteristic reasons, including (a) fraudsters constantly change their strategies, and accordingly, DNNs encounter observations that are not generated by the same process as the training distribution, (b) owing to the time-consuming process, very few transactions are timely checked by professional experts to update DNNs. Therefore, this study proposes three uncertainty quantification (UQ) techniques named Monte Carlo dropout, ensemble, and ensemble Monte Carlo dropout for card fraud detection applied on transaction data. Moreover, to evaluate the predictive uncertainty estimates, UQ confusion matrix and several performance metrics are utilized. Through experimental results, we show that the ensemble is more effective in capturing uncertainty corresponding to generated predictions. Additionally, we demonstrate that the proposed UQ methods provide extra insight to the point predictions, leading to elevate the fraud prevention process.

show abstract

Fraud Detection using Machine Learning and Deep Learning

Cited by 94 publications

References 4 publications

Pemetaan Bentuk Dan Pencegahan Penipuan E-Commerce

Pemetaan Bentuk Dan Pencegahan Penipuan E-Commerce

Karşılaştırmalı Veri Madenciliği Teknikleri Kullanılarak Bir E-Ticaret Sitesinin Kredi Kartı Dolandırıcılığı Sınıflandırması

Uncertainty-Aware Credit Card Fraud Detection Using Deep Learning

Contact Info

Product

Resources

About