Many businesses in the telecommunications sector are turning to Hadoop-based big data analytics solutions in order to tackle fraud, a new survey has found.

Research by Cloudera and Argyle Data noted that fraudulent activities are one of the biggest challenges for the industry, with telcos in the US alone losing around $38 billion a year in revenue to this.

Therefore, any technologies that these enterprises can put in place to help identify suspicious activity and put a stop to it before it becomes a major issue will be hugely valuable, and increasingly, Hadoop is seen as the answer.

Nine out of ten telcos (90 per cent) attending a recent webinar organised by Cloudera and Argyle Data stated they intend to use Hadoop to assist in their fraud prevention strategies.

However, just a third (34 per cent) said they currently have a platform in place for this, which indicates there is still a long way to go for the sector as a whole as they try to identify the best use cases for the technology.

Vijay Raja, solutions marketing manager at Cloudera, noted: "Fraud prevention is a textbook use case for Hadoop-based analytics because the ROI is immediately visible.  Real-time machine learning relies on large amounts of data to detect sophisticated revenue threats."

Platforms that are able to combine real-time analytics, machine learning and graphical visibility tools are essential in countering telecoms fraud. These solutions enable analysts to spot fraud attempts as they happen, which traditional monitoring systems can struggle to achieve.

As today's sophisticated, high volume attacks can cost communication service providers millions of dollars in revenue in a matter of minutes, being able to detect fraud quickly will be essential.

Arshak Navruzyan, vice-president of product management at Argyle Data, said: "Unsupervised machine learning delivers everything telco fraud analysts need to be efficient at and deliver immediate ROI."