Native Hadoop-based application leverages Cloudera Enterprise to deliver real-time fraud detection for mobile providers
Argyle Data, the leader in native Hadoop applications for revenue threat analytics in mobile communications, today announced the certification of its real-time, machine learning application suite with Cloudera Enterprise, the fastest, easiest, and most secure data management and analytics platform built on Apache Hadoop and the latest open source technologies.
“Communications fraud is a $38 billion dollar problem for communications service providers, according to the Communications Fraud Control Association,” said Dr. Ian Howells, Chief Marketing Officer at Argyle Data. “Argyle Data’s revenue threat analytics application suite is built from the ground up to run natively on Hadoop, enabling mobile operators to use machine learning on big data to detect previously undiscoverable fraud in minutes rather than hours or days, discover both new and old fraud attack patterns, and dramatically reduce false positives, saving millions of dollars.”
Argyle Data’s certification includes integration with Apache Impala (incubating) and support for both Apache Parquet and Apache Avro storage formats. Impala is a purpose-built, state-of-the-art analytic database architected specifically to leverage the flexibility and scalability of Hadoop for interactive business intelligence and data discovery.
“As part of the certification program we worked closely with Cloudera to provide a highly optimized integration designed to deliver peak levels of performance and availability, both on today’s architecture and future enhancements such as [Apache] Kudu (incubating),” said Howells. “Our goal was to allow our customers to focus their efforts to drive down fraud without being encumbered by technical roadblocks and integration challenges.”
Parquet is a column-oriented binary file format intended to be highly efficient for the types of large-scale queries that Impala is best at. Parquet is especially good for queries scanning particular columns within a table. Argyle Data is also working closely with Cloudera to integrate and leverage the benefits of Kudu in future releases.
Argyle Data will be showcasing how the world’s largest mobile operators are using Hadoop and machine learning to fight fraud at the Cloudera booth at Mobile World Congress, 6M30, Hall 6.