Künstliche Intelligenz Machine Learning Parts

How Machine Learning Is Preventing Fraud in E-commerce

E-commerce is continuing to grow at a breakneck pace, nearing $3 trillion worldwide in 2018, and increase of 18 percent from the previous year. Unfortunately, with that massive growth in sales has also come a significant growth in fraud. According to Experian, e-commerce fraud rose an astonishing 30 percent in 2017 (data for 2018 is unavailable at this time of writing). In other words, fraud grew at nearly twice the rate as actual purchases! And given the dollar volume of purchases in e-commerce, that growth represents losses in the billions.

So, what are e-commerce firms doing to fight back and protect themselves? Many are turning to machine learning, a subcategory of artificial intelligence, to help detect and thwart fraud before it occurs. Let’s take a closer look at machine learning and how it is being used to curb fraud—and ultimately save money.

What is machine learning?

Machine learning is essentially a type of artificial intelligence software that, instead of being programmed directly to do something, is programmed to learn from experience in order to improve outcomes. In other words, it is software that teaches itself how to do something by analyzing data, developing algorithms that describe a pattern, then test that algorithm against new data to determine its accuracy.

What does this look like in fraud prevention?

Well, a machine learning system can analyze all the transactions than an e-commerce seller has completed, including ones that turned out to be fraud. Then, based on that data, it can create an algorithm that predicts when a transaction is fraudulent. For example, it might develop an algorithm that says, “If an account adds three credit cards simultaneously, and two of them are rejected, then this has a high probability to be fraud.” That algorithm can then flag transaction as potential fraud when an account holder behaves like this, and the merchant can then take steps to verify the user’s identity.

What’s more, new data can be introduced at any time, which can help the system become smarter and more accurate. For example, it might detect a series of fraudulent transactions originating from a specific geographic location or for a specific amount. This is essential, as people looking to commit fraud are constantly changing their tactics, making machine learning an important tool for e-commerce merchants to stay on top of potential fraudulent transactions.

Combating identity theft in e-commerce

Another source of fraud is when an account gets broken into. Fraudsters will buy (or steal) account information on the black market, then use that to get into people’s accounts and make purchases using their information that is already on file.

A machine learning system can analyze data about normal customer behavior, such as what devices they typically sign in from, the time, the location, etc., and then use that to identify abnormal account activity.

By using advanced machine learning systems, merchants can deploy powerful fraud detection tools that will enable them to detect and prevent more fraud. And this comes just in time, as the methods and threats continue to multiply each year.

If you’re ready to explore ways machine learning and artificial intelligence can help your business protect itself, contact us. We are happy to help identify ways this cutting-edge technology can make your business smarter.

Best regards,
Florian Horn

0 replies

Leave a Reply

Want to join the discussion?
Feel free to contribute!

Leave a Reply

Your email address will not be published. Required fields are marked *