6 Ways Kount is Boosting Machine Learnings’ Ability To Fight Fraud

March 20, 2018

In today’s world, you would be hard-pressed to identify a successful organization that does not rely on data to make decisions in all aspects of business — and fighting fraud is no exception.  When discussing data, there are few industries that analyze more data than the fraud industry. 

Today’s fraud industry is based on historical, behavioral, and cognitive data that is collected at various points throughout the transaction process. For online merchants, data collection begins when the individual consumer signs onto a website, and then continues until the transaction is completed with an approved or disapproved rating.  The application of these advanced fraud-oriented analytics, applied to data from around the world (i.e. shared global data pool), typically takes a fraction of a second.  The amount of transactional data the fraud industry evaluates daily is enormous and only possible with the use of advanced machine learning.

This week, Kount will be exhibiting at MRC Vegas 2018, where we will be announcing a new feature based on Kount’s supervised machine learning called Boost™ Technology called the Boost™ Safety Rating.  This is a significant announcement for a number of reasons – the biggest being that customers will now be presented with a single numerical value describing the relative safety of a transaction indicating its legitimacy.  This is a look at the other side of a transaction, the legitimate transaction that merchant want to recognize quickly and allow acceptance. As an evolving application, Kount’s Boost Technology is able to assimilate data in real time to continually improve its ability to predict the risk of fraud for the individual merchant.

The Boost Safety Rating compliments Kount’s patented unsupervised machine learning, or Persona technology, to deliver merchants both supervised and unsupervised machine learning fraud detection and mitigation.  This combination of machine learning analysis provides Kount customers with the ability to simultaneously process millions of data points with each transaction to deliver individual merchants an enhanced confidence on the value of a digital transaction.  Specifically, Kount’s Boost Technology:

  • Assigns a single numeric value to predict fraud in real time.
  • Weighs the risk of fraud against the value of each unique customer to identify legitimate transactions from fraudulent ones.
  • Aggregates millions of transactions and their outcomes, including approvals, chargebacks, refunds, and reviews.
  • Calculates in real time, maintaining Kount’s current response time of roughly 250 milliseconds.
  • Harnesses data insight from billions of transactions that are part of Kount’s global network of merchants in nearly every vertical market.
  • Helps merchants protect good customers by measuring safety instead of just risk.

Because fraud solutions are focused on detecting, triaging, and building investigations on suspicious activity, it is critical that they have access to real-time data to respond promptly. Kount’s Boost Safety Rating provides a critical step to streamlining and allocating resources and mitigating fraud before it impacts an organization’s bottom line.

Unlike other industries, the fraud industry is ever changing. Today’s fraudsters don’t have to comply with rules and regulations when they are attacking merchants — they identify a tactic that works and look to exploit that strategy to achieve the greatest gains. This demands that organizations understand the evolving trends in real-time, while drilling down on the data and making more intelligent decisions.

Organizations that leverage machine learning are empowering their decision makers to access data, understand its meaning and make informed decisions to stop fraud before it impacts the businesses’ bottom line and the overall brand.

To learn more about the new Boost Safety Rating, a feature of Kount’s Boost  Technology, please visit the Kount news section of the website or visit Kount at MRC (Booth #709).