Skip to Main Content
Pinpoint suspicious financial transactions with Oracle Spatial and Python

About This Workshop

Youtube Video

About This Workshop
You've been alerted to possible fraudulent use of your credit card. How was it determined that fraud is likely? Fraud detection relies on the ability to identify outliers and anomalies compared to "normal" behavior. The analysis of location patterns plays an essential role in the process. For example, within a narrow time window is there a transaction far from a geographically concentrated set of typical transactions? Oracle Autonomous Database includes Oracle Spatial with features for geospatial data management and analysis. Combining Oracle Spatial with robust Python libraries for geospatial ML provides a robust platform to address the challenge of transaction fraud. In this workshop you will prepare data and perform feature extraction in Oracle Spatial and operate on that data with spatial ML algorithms in a Python notebook to ultimately find the needle of fraud in the haystack of financial transactions.

Workshop Info

1 hour
  • Lab 1: Prepare environment
  • Lab 2: Load, configure, explore  data
  • Lab 3: Perform feature extraction
  • Lab 4: Apply ML: models in Python to predict fraud
  • Familiarity with Database and Python is desirable, but not required
  • Some understanding of cloud and database terms is helpful
  • Familiarity with Oracle Cloud Infrastructure (OCI) is helpful

Other Workshops you might like