Signal processing on graphs for improving automatic credit card fraud detection

Vergara, Luis; Salazar, Addisson; Belda, Jordi; Safont, Gonzalo; Moral, Santiago; Iglesias, Sergio

doi:10.1109/ccst.2017.8167820

Cited by 23 publications

(12 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Finally, a future line of research is to apply novel methods developed from the signal processing on graphs framework to the problem proposed in this work. Recently, these methods have shown interesting results in data analysis for several applications that could complement the ones obtained by traditional statistical methods (see for instance, Belda et al 2017;Vergara et al 2017).…”

Section: Discussionmentioning

confidence: 99%

Effect of exploitation and exploration on the innovative as outcomes in entrepreneurial firms

et al. 2018

Self Cite

View full text Add to dashboard Cite

The main aim of this study is to establish the effect of the Exploitation and Exploration; and the influence of these learning flows on the Innovative Outcome (IO). The Innovative Outcome refers to new products, services, processes (or improvements) that the organization has obtained as a result of an innovative process. For this purpose, a relationship model is defined, which is empirically contrasted, and can explains and predicts the cyclical dynamization of learning flows on innovative outcome in knowledge intensive firms. The quantitative test for this model use the data from entrepreneurial firms biotechnology sector. The statistical analysis applies a method based on variance using Partial Least Squares (PLS). Research results confirm the hypotheses, that is, they show a positive dynamic effect between the Exploration and the Innovative as outcomes. In the same vein, they results confirm the presence of the cyclic movement of innovative outcome with the Exploitation.

show abstract

Section: Discussionmentioning

confidence: 99%

Effect of exploitation and exploration on the innovative as outcomes in entrepreneurial firms

et al. 2018

Self Cite

View full text Add to dashboard Cite

show abstract

“…where J is the number of scales, N is the length of the index curve, and S J ̃ and D j ̃, respectively, denote the tendency and local details of the index curve v at each scale. The proposed augmentation strategy adopts a stochastic version of IAAFT algorithm [54], [55] to each set of the detailed coefficients independently to obtain randomized values that retain the original values and periodicities. It should be mentioned that both the amplitude spectrum and phase spectrum are perturbed at each level of MODWT coefficients.…”

Section: B Wavelet Transform-based Augmentation Of the Index Curves P...mentioning

confidence: 99%

“…Recently, Fourier transform (FT) based data augmentation approaches, such as Iterative Amplitude Adjusted Fourier Transform (IAAFT), adopted phase alternations to preserve the power-and cross-spectrum of the signals [52]- [54]. The stochastic versions of IAAFT, which implement fractional amplitude adjustment, preserve the amplitude distribution and the power spectrum of the measured time series [55]. Kayal et al [56] employed jittering of discrete cosine and wavelet transforms' bases to generate surrogates having slightly different noise properties.…”

mentioning

confidence: 99%

Augmentation of Vegetation Index Curves Considering the Crop-Specific Phenological Characteristics

Arun

Karnieli

2022

IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing

View full text Add to dashboard Cite

The state-of-the-art crop phenological classifiers use vegetation index (VI) time-series data and deep learning (DL) techniques. However, the scarcity of training samples limits the performance of these approaches. Unlike the conventional augmentation techniques, the data augmentation of VI curves should preserve the crop-specific phenological events. The DLbased augmentation approaches do not give good results when the training samples are limited. Also, the conventional approaches such as translation, rotation, scaling, and wrapping do not preserve the characteristic features of the index curves, thereby making them inappropriate for the VI-curve-based augmentations. This study proposes a non-DL-based data augmentation strategy that requires only a minimal number of actual training samples. In the proposed approach, the periodic phenological events and the underlying trend for each crop class are modeled to improve the augmentation. The trends of different crop classes are estimated by jointly maximizing the autocorrelation and variance, while the optimal subsequences are generalized as the phenological events. The proposed augmentation strategy of using Maximal Overlap Discrete Wavelet Transform (MODWT) for obtaining the surrogates that retain the crop-specific features and periodicities significantly improves the results. It may be noted that the proposed approach does not alter the wavelet coefficients that are characteristics of a given crop class. The experiments using time series VI data, covering 90 fields of wheat and 60 fields of barley, confirm better accuracy of the proposed augmentation approaches as compared to the prominent approaches.

show abstract

“…In future, advanced concepts of fuzzy will be utilized so that fraud detection can be identified easily. [18] Presented in order to enhance the performance of credit card fraud detection systems, they proposed various methods which are based on signal processing on graphs. In this paper, they proposed iterative amplitude adjusted Fourier transform along with iterative surrogate signals on graph algorithms as an alternative.…”

Section: Literature Reviewmentioning

confidence: 99%

Fraud prediction for credit card using classification method

Monika

Kaur

2018

IJET

View full text Add to dashboard Cite

With the improvement of innovation like credit cards, debit cards, mobile banking, Internet managing an account is the mainstream medium to exchange the cash starting with one record then onto the next. Credit card is picking up fame day by day which expands the online exchange with the expansion in online shopping, online charge payment, insurance premium and different charges so the extortion cases identified with this are likewise expanding and it puts an extraordinary anxiety on the economy, affecting the two clients and budgetary bodies. It costs cash as well as an awesome measure of time to reestablish the damage done. In this paper, we look whether data mining procedures are valuable to estimate and categorize the client's credit risk score (normal/fraud) to beat the future dangers. The reason for existing is to keep the clients from online exchange by utilizing particular Data mining classification methods. The fakes are ascertained by Naïve Bayes method way to deal with break down the exchange is actual or fake. The exploratory outcome demonstrates that our model has great classification accuracy, recall and precision.

show abstract

Signal processing on graphs for improving automatic credit card fraud detection

Cited by 23 publications

References 26 publications

Effect of exploitation and exploration on the innovative as outcomes in entrepreneurial firms

Effect of exploitation and exploration on the innovative as outcomes in entrepreneurial firms

Augmentation of Vegetation Index Curves Considering the Crop-Specific Phenological Characteristics

Fraud prediction for credit card using classification method

Contact Info

Product

Resources

About