Answers are grounded in this paper and cite their sources.
In the real world, online transactions are increasing rapidly to facilitate the users with better comfort services. At present, it is essential to increase privacy to make the digital world safe and secure and to analyze transactional data timely to prevent transactional fraud. The financial fraud can be prevented by detection and prediction from transactional data to secure future transactions. Various techniques are available for financial fraud detection. However, the major issue is analyzing the user's real data which is too risky and difficult to access. The organizations do not have any right to provide their customers with data due to authenticity and privacy concerns. Therefore, it required to generate the synthetic temporal data. This paper proposes a model for fraud detection with synthetic data generation using GANs. We have illustrated the model on synthetic data to perform the experiment. Results obtained from experiment shows that the proposed model outperforms with synthetic data and visualize the reliability of the generated synthetic Metadata over original data. The proposed model provides the facility to deal with less data availability by providing the best solution on transactional data for fraud detection.
No takes yet. Share an insight, caveat, or question.
Jaiswal et al. (2022) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: