Table 1: Review of financial fraud detection methods.

Cited UsedData DescriptionApplied 

[27]Credit and debit card transactions of the Spanish bank BBVA, from January 2011 to December 2012Multilayer perceptron (MLP)True Positive Rate (TPR), Receiver Operating Characteris-tic (ROC curves)The improvement of accuracy in detecting results by using parenclitic networks reconstruction for feature extractionRequires the cases where the features are not correlated or not extracted by parenclitic networks
[28]Transactions of National banking group of Italy, from April 2013 to August 2013Multi-objective genetic algorithmTPR, ROC
Provide feature selection process via the auto-tuning methodShould be applied to cases other than Banksealer
[29]UCSD Data Mining Contest 2009 dataDeep neural
network (DNN)
Mean Squared Error (MSE),
Root Mean Squared Error (RMSE),
Mean Absolute Errors (MAE),
Root Mean Squared Log Error (RMSLE)
Study on the importance of features based on the deep learning methodDoes not have accurate experimental explanation process and validation
[30]Dataset achieved from the second robotic & artificial intelligence festival of Amirkabir UniversityDecision treesF-Measure
classification is performed using cost-sensitive decision trees in a decision forest framework
Having a class imbalance problem
[31]German dataset (which has been used in KDD99
competition), Australian credit cards’ open dataset
Particle swarm optimization (PSO), Teaching-learning- based optimization (TLBO)Confusion Matrix (True positive, True negative, False positive, False negative)Experiment with various datasets
Detection accuracy is relatively low
[32]Not specificLinear Regression,
Artificial Neural Networks (ANN),
K-Nearest Neighbor (KNN), Support Vector Machine (SVM), Decision Stump, M5P Tree, Decision Table
Normalized Root Mean Squared Error (NRMSE), TPR,
A comparative study using various algorithmsDetection accuracy should be increased
[33]UCI German
credit card dataset
Cross validation
As Gaussian kernels including RBF are with appropriate regularization, it guarantees a globally optimal predictor which minimizes both the estimation and approximation errors of a classifierThere is no comparison with other algorithms and there is no explanation to verify SVM algorithms superiority to others
[34]Credit card transaction data from commercial bank in ChinaConvolutional Neural Networks (CNN),
F-MeasureDesigning a feature called trading entropy based on the latest consumption preferences for each customer and generating synthetic fraudulent samples from real frauds by a cost-based sampling methodDetection
accuracy should be increased
[35]Banking transaction dataset in IranKNNAccuracy,
A novel approach combining K-nearest neighbor, association rules like Apriori algorithmThe validation is not specific and it is difficult to compare the proposed results with other algorithms
[36]Open dataset: ccFraudNN, PSO,
Auto-associative neural network (AANN), Particle swarm optimization auto-associative neural network (PSOAANN)
MSE, Classification Rate (CR)Combined parallelization of the auto-associative neural network in the hybrid architectureDataset is highly unbalanced and detection accuracy should be increased
[37]Open dataset:
MIT Human Dynamics Lab
SVM, Fuzzy clusteringTPR, FPR, ROC
Divide the fraud detection system into two principal modulesWould be better to compare it with more diverse algorithms
[38]Transactions from a large national bank in Italy,
collected from December 2012 to August 2013
Principal component analysis (PCA),
Operate in online processingAccuracy by validation is not constant
[39]Not specificSelf- organizing map (SOM)TPR, FPRDivision of transactions to form an input matrix and ability to be applied to a large complex setOnly have compared to one algorithm