Reproduce “Improving Financial Cybersecurity Through Fraud Transaction Detection with Deep Learning” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Harish Kant Pathak, Rahul Kumar
The authors evaluate a wide-and-deep neural network for fraud classification on one transaction dataset. It combines a linear route for simpler feature relationships with a deeper route for more complex patterns. They report higher accuracy, precision, and F1 than an ANN, but lower recall and tied ROC-AUC. Conflicting confusion-matrix counts and unclear data preparation limit interpretation and broader claims.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.