How to use analytics in the fight against fraud

Written by on August 3, 2017 in Guest Blog with 0 Comments

The telecoms sector is expanding its services around the world with faster connections and more connected devices. However, new services create not just new revenue opportunities but also associated risks for operators. The challenge is to secure the upside, while at the same time, minimise potential losses incurred via fraudulent activity.

One of the main problems with modern fraud techniques is that operators can’t keep up. Fraudsters know all too well that the majority of solutions are reactive rather than proactive, and are effectively exploiting this to increase the time with which it takes to discover their new schemes. This has resulted in fraud methods becoming ever more complex and sophisticated.

And with the continually evolving and increasingly complex nature of fraud attacks only set to advance further, CSPs are under increased pressure to bridge the gap between providing a high level of customer service and preventing fraud. All too often a fraud isn’t actually even discovered until after a CSP’s bottom line has been severely impacted.

So how can operators keep up? Most believe the answer lies in analytics. CSPs need a comprehensive, multi-protocol solution that is quick, nimble and adds to an operator’s existing system capabilities. Effective defence requires a number of key elements:

Capturing the appropriate data

Filtering and blocking abnormal and suspicious traffic or activities

Advanced analytical methods to stop sophisticated fraud

 The leaps made in Big Data and Analytics in the past few years means real-time, advanced methods of pattern identification can be designed and operated by experienced analytics and fraud professionals. In other words, analytics is no longer a large dataset to be assessed after the fact.

CSPs must start by using defensive analytics to monitor traffic at all interconnecting points – domestic and international – capturing data for both inbound and outbound roamers using in-signalling nodes.

As well as constantly analysing traffic from the various relevant locations, an analytics system is required to filter abnormal and suspicious traffic based on pre-configured rules. Using machine learning outlier detection also enables it to go beyond the rule-based systems and catch the fraud threats at an earlier stage.

Using analytics against fraudsters, CSPs will be able to access many benefits, including:

  • Simple access – delivery of fraud warnings in clear, easy-to-understand language and processes
  • Accurate insights – the best solutions provide a delicate balance between missing a few genuine instances of fraud (false negatives), while at the same time not identifying too many false positives
  • Active testing – an agent that is completely configurable and can be scheduled to generate test call results
  • Data science – advanced machine learning capabilities utilise supervised and unsupervised machine learning techniques to detect various known and unknown fraud scenarios

 The high degree of network complexity and ever evolving use of sophisticated techniques is making fraud increasingly more difficult to detect and prevent. However, with these advancements, CSPs can and advance their fraud protection tactics. We believe the use of analytics is the best way to fight new fraud in real-time and, ultimately, stop it in its tracks.

By Jason Lane-Sellers, Offering Manager, Fraud and Security, Mobileum

Tags: , , ,

Guest Writer

About the Author

About the Author: This article has been supplied by one of our many guest writers that have specialist interest in the topic featured. .

Subscribe

If you enjoyed this article, subscribe now to receive more just like it.

Subscribe via RSS Feed

Leave a Reply

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

Top
%d bloggers like this: