Credit Card Fraud Detection - Machine Learning methods

Varmedja, Dejan; Karanovic, Mirjana; Sladojević, Srdjan; Arsenović, Marko; Anderla, Andraš

doi:10.1109/infoteh.2019.8717766

Cited by 186 publications

(81 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The extreme imbalance ratio of 577:1 is incomparable to any of the datasets in imblearn.datasets. Also, this dataset has received special attention of researchers attempting to use ML in Credit fraud detection (Varmedja et al 2019). In this article we see that lr and rf have good prediction accuracies on the dataset.…”

Section: Methodsmentioning

confidence: 78%

“…Thus we chose these two ML models for the credit fraud dataset. Varmedja et al (2019) has also not provided cross validated analysis of their models, while our models have been trained and tested with the usual tenfold cross validation framework as discussed before. Also, for two small datasets with a critically small minority class, we used only knn and lr classifiers, with parameter settings as specified before.…”

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

LoRAS: an oversampling approach for imbalanced datasets

et al. 2020

View full text Add to dashboard Cite

The Synthetic Minority Oversampling TEchnique (SMOTE) is widely-used for the analysis of imbalanced datasets. It is known that SMOTE frequently over-generalizes the minority class, leading to misclassifications for the majority class, and effecting the overall balance of the model. In this article, we present an approach that overcomes this limitation of SMOTE, employing Localized Random Affine Shadowsampling (LoRAS) to oversample from an approximated data manifold of the minority class. We benchmarked our algorithm with 14 publicly available imbalanced datasets using three different Machine Learning (ML) algorithms and compared the performance of LoRAS, SMOTE and several SMOTE extensions that share the concept of using convex combinations of minority class data points for oversampling with LoRAS. We observed that LoRAS, on average generates better ML models in terms of F1-Score and Balanced accuracy. Another key observation is that while most of the extensions of SMOTE we have tested, improve the F1-Score with respect to SMOTE on an average, they compromise on the Balanced accuracy of a classification model. LoRAS on the contrary, improves both F1 Score and the Balanced accuracy thus produces better classification models. Moreover, to explain the success of the algorithm, we have constructed a mathematical framework to prove that LoRAS oversampling technique provides a better estimate for the mean of the underlying local data distribution of the minority class data space.

show abstract

Section: Methodsmentioning

confidence: 78%

Section: Methodsmentioning

confidence: 99%

LoRAS: an oversampling approach for imbalanced datasets

et al. 2020

View full text Add to dashboard Cite

show abstract

“…Unsupervised learning outperforms supervised methods when considering skewness [19]. However, oversampling methods like SMOTE have proven [41] to improve accuracy in supervised models. Lastly, recent research concludes that combining multiple outlier scores can negatively impact accuracy [30].…”

Section: ) Unsupervised Techniquesmentioning

confidence: 99%

Evaluation of Deep Neural Networks for Reduction of Credit Card Fraud Alerts

Carrasco

Sicilia-Urbán

2020

IEEE Access

View full text Add to dashboard Cite

Fraud detection systems support advanced detection techniques based on complex rules, statistical modelling and machine learning. However, alerts triggered by these systems still require expert judgement to either confirm a fraud case or discard a false positive. Reducing the number of false positives that fraud analysts investigate, by automating their detection with computer-assisted techniques, can lead to significant cost efficiencies. Alert reduction has been achieved with different techniques in related fields like intrusion detection. Furthermore, deep learning has been used to accomplish this task in other fields. In our paper, a set of deep neural networks have been tested to measure their ability to detect false positives, by processing alerts triggered by a fraud detection system. The performance achieved by each neural network setting is presented and discussed. The optimal setting allowed to capture 91.79% of total fraud cases with 35.16% less alerts. Obtained alert reduction rate would entail a significant reduction in cost of human labor, because alerts classified as false positives by the neural network wouldn't require human inspection.

show abstract

“…Em [Varmedja et al 2019] também foi utilizado o mesmo conjunto de dados. Foi aplicada a técnica SMOTE para sobreamostragem e feita uma seleção de atributos para comparar os algoritmos de regressão logística, Naïve Bayes, random forest e multilayer perceptron.…”

Section: Trabalhos Correlatosunclassified

“…Por exemplo, em [Khatri et al 2020]é realizado um estudo comparativo de diferentes técnicas de aprendizado de máquina, porém sob o conjunto original desbalanceado. Já em [Sahin and Duman 2011], [Dhankhad et al 2018], [Mishra and Ghorpade 2018], [Niu et al 2019] e [Varmedja et al 2019], utiliza-se umaúnica técnica de balanceamento, enquanto que em [Awoyemi et al 2017]é utilizada uma combinação de duas técnicas.…”

Section: Introductionunclassified

Avaliação empírica de classificadores e métodos de balanceamento para detecção de fraudes em transações com cartões de créditos

Nicola¹,

Lauretto²,

Delgado³

2020

Anais Do Encontro Nacional De Inteligência Artificial E Computacional (ENIAC 2020)

View full text Add to dashboard Cite

Algoritmos de aprendizado de máquina são amplamente utilizados em sistemas para detecção de fraudes em cartões de crédito devido à capacidade de distinguir entre transações legítimas e fraudulentas. Um problema reconhecido nesta área é o alto desbalanceamento usualmente encontrado nas classes, que pode comprometer o desempenho dos classificadores. Os estudos empíricos encontrados na literatura aplicam, no máximo, duas técnicas de amostragem. Este artigo traz um estudo comparativo de cinco modelos de classificação sob cinco diferentes métodos de balanceamento dos conjuntos de treinamento. O melhor desempenho foi obtido pela random forest, que além de ter apresentado o maior F-score médio (0,867), mostrou-se consideravelmente mais robusta do que os demais classificadores em relação à escolha da técnica de balanceamento e à seleção de atributos.

show abstract

Credit Card Fraud Detection - Machine Learning methods

Cited by 186 publications

References 11 publications

LoRAS: an oversampling approach for imbalanced datasets

LoRAS: an oversampling approach for imbalanced datasets

Evaluation of Deep Neural Networks for Reduction of Credit Card Fraud Alerts

Avaliação empírica de classificadores e métodos de balanceamento para detecção de fraudes em transações com cartões de créditos

Contact Info

Product

Resources

About