New Azure Architecture - Detect mobile bank fraud

In a typical case of online fraud, the thief makes multiple transactions, leading to a loss of thousands of dollars. That's why fraud detection must happen in near real-time. 

 

This article presents a solution that uses Azure technology to predict a fraudulent mobile bank transaction within two seconds. We've built it with customers.

 

Read the article here:

• Detect mobile bank fraud - Azure Architecture Center

 

Let's dig into the architecture:

 

 

An event-driven pipeline ingests and processes log data, creates and maintains behavioral account profiles, incorporates a fraud classification model, and produces a predictive score. Most steps in this pipeline start with an Azure function. A model training workstream combines on-premises historical fraud data and ingested log data. Azure Data Factory orchestrates the processing steps. We use Azure Logic Apps to connect and synchronize to an on-premises system to create a fraud management case, suspend account access, and to generate a phone contact.

 

In the article you'll find:

• Information about the top challenges: Rare instances of fraud and rigid rules.
• Operational context: The key questions we asked and how fraud is committed in the operational environment.
• Compromise matrix: See the methods used, data taken, and patterns for several types of fraud, including Credential, Device, Financial, and Non-Transactional compromises.
• A detailed dataflow of the above architecture.
• Data pipeline and automation: What happens in the two seconds, in order to catch the compromise.
• Event processing: Architecture and dataflow that explains in detail the fundamental interactions for an Azure function within this infrastructure.
• Data pre-processing and JSON transformation.
• Near real-time data processing and featurization with SQL Database.
• Event schema management.
• Feature engineering for machine learning.
• AutoML: It automates the time-consuming, iterative tasks of machine learning model development.
• Data imbalance: In a fraud dataset, there are many more non-fraudulent transactions than fraudulent transactions.
• Model training with a code sample!
• Model evaluation: The account-level metrics are described in a table.
• Model operationalization and retraining.
• Components: Direct links to all the Azure services used in this solution.
• Technical considerations: Skill sets and Hybrid operational environment.
• Security considerations: Includes a Networking Security Architecture and a security baseline recommendations matrix.
• Scalability considerations.

 

You can find the article here, on the Azure Architecture Center:

 

• Detect mobile bank fraud - Azure Architecture Center

 

 

Special thanks to the Engineers who wrote this:
- Kate Baroni
- Michael Hlobil
- Cedric Labuschagne
- Frank Garofalo
- Shep Sheppard

And thanks also to our editor/tech writer, Mick Alberts.

 

Remember to keep your head in the Cloud!

 

Ed