Lina Ni

This paper proposes a C-RNN forecasting method for Forex time series data based on deep-Recurrent Neural Network (RNN) and deep Convolutional Neural Network (CNN), which can further improve the prediction accuracy of deep learning algorithm for the time series data of exchange rate. We fully exploit the spatio-temporal characteristics of forex time series data based on the data-driven method. On the exchange rate data of nine major foreign exchange currencies, the experimental comparison of the forecasting method shows that the C-RNN foreign exchange time series data prediction method constructed in this paper has better applicability and higher accuracy.

With the flourishing of the credit card business and Internet technology, the risk of fraudulent credit card transactions is ever-increasing due to the complex information involved in the credit card business. Since the high redundancy of feature information and imbalance of class distribution in transaction data, the performance of the existing machine learning-based models for detecting credit card fraudulent transactions still needs to be improved. Therefore, it is crucial to build fraud detection models for effective feature engineering and sampling techniques. This article proposes a credit card fraud detection model incorporating a fraud feature-boosting mechanism with a spiral oversampling balancing technique (SOBT). Specifically, we present a compound grouping elimination strategy to exclude highly redundant and correlated features from the credit card transaction dataset and improve the data quality. Furthermore, we design a multifactor synchronous embedding mechanism, which combines the performance evaluation metrics of the embedding model for each feature and improves the decision-making ability of each feature for the target domain. Moreover, we propose an SOBT to balance the ratio of legitimate to fraudulent transactions, which improves the ability of the fraud detection model to distinguish legitimate from fraudulent transactions. Extensive experimental results based on two real-world datasets demonstrate that our methods can facilitate efficient credit card fraud detection and achieve better performance than state-of-the-art methods.

The proliferation of ubiquitous Internet and mobile devices has brought about the exponential growth of individual data in big data era. The network user data has been confronted with serious privacy concerns for extracting valuable information during the process of data mining. Differential privacy preservation is a new paradigm independent of the adversaries' prior knowledge, which protects sensitive data while maintaining certain statistical properties by adding random noise. In this paper, we put forward a differential privacy preservation multiple cores DBSCAN clustering schema based on the powerful differential privacy and DBSCAN algorithm for network user data to effectively leverage the privacy leakage issue in the process of data mining, enhancing data clustering efficaciously by adding Laplace noise. We perform extensive theoretical analysis and simulations to evaluate our schema and the results show better efficiency, accuracy, and privacy preservation effect than previous schemas.