pgabriel-01 / fraud_detection_with_spark

Detecting fraud with Azure Machine Learning Service, Synapse Spark Pool, and PySpark

Fraud_Detection_with_Spark

Detecting fraud with Azure Machine Learning Service, Synapse Spark Pool, and PySpark

Synopsis:

Why is detecting fraud so important?

• Takes away resources from legit customers.
• Affects getting a clear picture of product and metrics.

Goal:

• To create customer core hour usage profile.
• Do anomaly detection to identify fraudulent customers.

Reference links:

• Kaggle
• Microsoft
• TorwardsDataScience

Submit Spark jobs in Azure Machine Learning (preview) [link]

Interactive Data Wrangling with Apache Spark in Azure Machine Learning (preview) [link]

• Import and wrangle data from Azure Machine Learning Datastore [link]
• Import and wrangle data from Azure Blob storage [link]
• Import and wrangle data from Azure Data Lake Storage (ADLS) Gen 2 [link]
• Accessing data on the default file share [link]

Identity used to access data

• Credential [link]
• Credential-less [link]