Fintech applications of Graph Network: Anti-fraud, Loan App Classification, ID-Mapping of Natural Person
Kai Zhang, Head of AI Lab, FinTell, Inc.
FinTell's AI Lab developed a series of graph learning models that applied in fintech and financial risk management, for instance, anti-fraud models for group fraud and propagated fraud, loan app classification, ID-mapping of natural persons. The models leveraged TigerGraph’s graph engine and a series of bipartite graph frameworks to aggregate modeling features from historical and spatial dimensions. Compared with the baseline model, the graph-based method demonstrated a boost for model discrimination power by the graph learning method and capability of TigerGraph. Business impact in financial risk management will be addressed as well.
Download the session slides.
Connect with Kia on LinkedIn