Mobile security is one of the major problems of modern life, considering that today, more than 800 million people use applications every day on their smartphones and tablets, which includes financial proposals such as banks and stores.
Thinking about it, Microsoft decided to make it available and to explain how to create and deploy an instance in Azure with artificial intelligence, capable of identifying and preventing fraud in less than two seconds.
According to the company, part of the fraud occurs through the phone number hijacking, known as “SIM swap attack”, thus receiving all the connections and SMS that originally had to arrive at the correct smartphone. This technique was even used by the American student who managed to steal $5 million from 40 people.
In order to reduce the impact of this type of attack, the fraud detection tool created by Microsoft, which is similar to that used by banks today, creates a user behavioral profile, evaluating the transactions and defining what action will be taken.
Although the system is very similar to what already exists in the market, its main differential is due to the speed of the process.
Behind the architecture, three main components of the platform belonging to Microsoft Azure (Functions, SQL and Machine Learning) are used, as well as the basis on three factors:
- Engineering resources to create customer profiles and accounts.
- Azure Machine Learning to create a fraud classification template.
- Azure PaaS services for real-time event processing and end-to-end workflow.
Also available from the leading developer, Kate Baroni, is a “Solution Guide: Mobile Banking Fraud” that delves deeper into the problems and solutions proposed within Azure in partnership with artificial intelligence.