The modern economy has seen a mass migration of retailers to e-commerce and other digital platforms. The cost and convenience advantages have produced a sales revolution for a number of varied industries but have also created new vectors for fraud and abuse. The semi-anonymous nature of internet-based transactions and digital platforms has provided fertile ground for scams, theft, and otherwise unseemly behavior.
It is into this brave new world of cybercrime that businesses hope that the latest big data platforms can make a positive difference. Leveraging powerful analytics tools and harnessing the massive quantities of data produced each day, companies hope to uncover potential fraud as it occurs. The most optimistic believe that the predictive capabilities of big data-driven artificial intelligence may someday end online crime altogether. That may well come to pass, but it will be an uphill climb.
The current state of affairs
The internet represents an ever-larger portion of global retail sales each year. Estimates indicate that e-commerce transactions alone will reach $2.3 trillion dollars this year. Digital platforms also generate enormous amounts of advertising revenue and related business. It’s an inviting target for fraudsters and cybercriminals, and they have done their best to exploit it. Current fraud statistics indicate that:
- $21.8 billion in global losses due to credit card fraud in 2015, with large recent increases in online credit card transaction fraud.
- $6.5 billion in advertising fee theft generated by fraudulent clicks and automated views.
- Global businesses lose a staggering $3.5 trillion to fraud of all types each year.
- An average of 8% of revenue spent on fraud prevention measures each year by businesses.
The sheer scale of the transactions, as well as the amount of losses, clearly demands a solution that only a big data platform can provide.
Machine learning and validity verification
There are already several examples of big data analytics being applied to combat digital fraud. Some of them are purpose-built all-in-one solutions, while others aim to unify and analyze the data gleaned from existing fraud detection systems. For example, analytics firm Splunk has created an anti-fraud system that leverages existing detection technology and applies machine learning, analysis and reporting tools to provide advanced early warning and reporting capabilities.
There are also startups that are attacking the problem from the consumer’s perspective. In the diamond retail industry, Rare Carat has created a system built on the IBM Watson platform to reduce fraudulent diamond sales. It’s an industry that’s been riddled with fraud for years, with any number of scams designed to mislead consumers and even other businesses.
The long-term outlook
In the digital marketplace, big data solutions have already helped to increase user engagement and sales. It’s not outside of the realm of expectation that they will also see success in fraud prevention applications. Judging by the solutions that have been fielded thus far, the future of digital fraud detection and prevention looks bright. It likely won’t be long until the tables are turned, and it will be the cybercriminals that end up on the losing edge of the technological fight.