/fraud-orchestration

Preempt fraud with rule-based patterns and select ML algorithms for reliable fraud detection. Use anomaly detection and fraud prediction to respond to bad actors rapidly.

Primary LanguagePythonOtherNOASSERTION

Databricks fraud framework

The financial service industry (FSI) is rushing towards transformational change to support new channels and services, delivering transactional features and facilitating payments through new digital channels to remain competitive. Unfortunately, the speed and convenience that these capabilities afford is a benefit to consumers and fraudsters alike. Building a fraud framework often goes beyond just creating a highly accurate machine learning model due ever changing landscape and customer expectation. Oftentimes it involves a complex decision science setup which combines rules engine with a need for a robust and scalable machine learning platform. In this series of notebook, we'll be demonstrating how Delta Lake, MLFlow and a unified analytics platform can help organisations combat fraud more efficiently

  • STAGE1: Integrating rule based with ML
  • STAGE2: Building a fraud detection model

© 2021 Databricks, Inc. All rights reserved. The source in this notebook is provided subject to the Databricks License [https://databricks.com/db-license-source]. All included or referenced third party libraries are subject to the licenses set forth below.

library description license source
shap Model explainability MIT https://github.com/slundberg/shap
networkx Graph toolkit BSD https://github.com/networkx
xgboost Gradient Boosting lib. Apache2 https://github.com/dmlc/xgboost
graphviz Network visualization MIT https://github.com/xflr6/graphviz
pandasql SQL syntax on pandas MIT https://github.com/yhat/pandasql/
pydot Network visualization MIT https://github.com/pydot/pydot
pygraphviz Network visualization BSD https://pygraphviz.github.io/