Detecting Fraudulent Bank Account Based on Convolutional Neural Network with Heterogeneous Data

Abstract

Detecting fraudulent accounts by using their transaction networks is helpful for proactively preventing illegal transactions in financial scenarios. In this paper, three convolutional neural network models, i.e., NTD-CNN, TTD-CNN, and HDF-CNN, are created to identify whether a bank account is fraudulent. The three models, same in model structure, are different in types of the input features. Firstly, we embed the bank accounts' historical trading records into a general directed and weighted transaction network. And then, a DirectedWalk algorithm is proposed for learning an account's network vector. DirectedWalk learns social representations of a network's vertices, by modeling a stream of directed and time-related trading paths. The local topological feature, generating by accounts' network vector, is taken as input of NTD-CNN, and TTD-CNN takes time series transaction feature as input. Finally, the two kinds of heterogeneous data, being integrated into a novel feature matrix, are fed into HDF-CNN for classifying bank accounts. The experimental results, conducted on a real bank transaction dataset, show the advantage of HDF-CNN over the existing methods. 2Mathematical Problems in Engineering of CNN structure makes it successful to address many classification problems. In a specific classification scenario, one can tune the structural feature settings of CNN, e.g., the layer numbers, the neuron numbers of each layer, the types of pooling functions, and activation functions, to achieve the best performance.Having abstracted the bank accounts into vertices and their transaction relationships into directed edges, the trading behaviors of accounts can be formed into a directed and weighted network. The transaction relationship information and time series information of bank accounts are embedded into the generated network. As mentioned above, CNN models obtain excellent performance in time series classification and social network. The superiority in convolution kernel and structural design inspires us to employ CNN framework in FAD issue. Therefore, with the labelled data provided by economic investigation experts, three convolutional neural network (CNN) models are proposed to address the FAD issue. The models are listed as follows.(1) A CNN model uses network topological data (NTD) being called NTD-CNN model.(2) A CNN model utilizes time series data (TTD) being referred to as TTD-CNN model. (3) A CNN model employs the two kinds of heterogenous data features (HDF), which are extracted from the former two kinds of data, being short for HDF-CNN model. The experiments on a real dataset, containing illegal pyramid selling accounts, demonstrate the effectiveness of our three CNN models. Except for the TTD-CNN, the other two CNN models achieve better performance than traditional abnormal detection method regarding precision, sensitivity, and F1-score. In summary, the classification performance of HDF-CNN is much better than that of the other three methods. To the best of our knowledge, this is the first time that CNN is ap...

Predicting Visual Outcome After Surgery in Patients With Idiopathic Epiretinal Membrane Using a Novel Convolutional Neural Network

“…Previously reported data fusion CNN models commonly use a matrix format to perform heterogeneous data fusion. 18,19 Hence, all heterogeneous data are merged into a single input and finally extracted through CNN. However, loss of information may occur in numeric data after convolutions and pooling.…”

Intelligent Analysis Strategy of Pragmatic Failure in Cross-Cultural Communication Based on Convolution Neural Network

“…Nongmeikapam et al added additional data sets by the means of data set expansion, which was well verified on handwritten digital picture data sets by using convolutional neural networks and improved the experimental performance when the data sets were insufficient [28]. Lv et al obtained three new network structures by modifying the activation function, learning rate, and changing the number of filters in the original network structure, namely CNN1-1, CNN1-2, and CNN1-3 [18]. Akhtar et al use a large amount of unlabeled audio data to learn features by using deep convolution belief network, and apply the learned features to specific speech recognition tasks and music recognition tasks [29].…”

“…e result is the nearest sample between the two. rough this means, we can get a good effect on the processing of some isolated words, but there are great defects in continuous speech, which does not fundamentally solve the problem of recognition [18]. Huang, Qian, and Zhu proposed to expand the research scope to the morphology of language.…”

Parallel path detection for fraudulent accounts in banks based on graph analysis

“…In the banking sector, machine learning ( Lv et al, 2019 ; Hashemi, Mirtaheri & Shamsi, 2021 ; Patil, Nemade & Soni, 2018 ) has emerged as a potent tool for detecting fraudulent accounts. Notably, graph neural networks ( Zeng & Tang, 2021 ; Xiang et al, 2023 ) excel in terms of detection accuracy, thanks to their ability to uncover intricate relationships and patterns within vast transaction datasets.…”