Digital payment fraud detection methods in digital ages and Industry 4.0

The advent of the digital economy and Industry 4.0 enables financial organizations to adapt their processes mitigate risks losses associated with fraud. Machine learning algorithms facilitate effective predictive models for fraud detection 4.0. This study aims identify an efficient stable model platforms be adapted By leveraging a real credit card transaction dataset, this proposes compares five different models: logistic regression, decision tree, k-nearest neighbors, random forest, autoencoder. Results show that forest regression outperform other algorithms. Besides, undersampling method feature reduction using principal component analysis could enhance results proposed models. outcomes studies positively ascertain effectiveness features selection sampling methods tackling business problems in new age industrial detect fraudulent activities.