Reliably classify thousands of messages as they come in The Messaging Classification Appliance processes any kind of unstructured text messages in real time with one of the highest precision levels on the market.

In a nutshell

  • High-performance message filtering and routing appliance
  • Exceptional throughput
  • Real-time processing
  • Human-level accuracy
  • Autoconfiguration: no manual tuning, fast training
  • Available in Q1 2020

In partnership with Xilinx and Supermicro, introduces its Messaging Classification Appliance, a pre-packaged software and hardware solution that can filter, classify, and route stream of messages in real time at a massive scale.

For example, the solution is able to route incoming inquiries from a general email address to the appropriate department based on the content of the email. The solution can be implemented to track customer satisfaction by extracting meaningful content from incoming messages, monitor outgoing emails for compliance adherence, or filter streams of news or social media data.

With's approach to messaging classification, enterprises have an out-of-the-box solution for mining emails for business content and classifying them, that comes as a pre-trained appliance with standard filters. If customized filtering is needed, it can be done with a fraction of the training data, as well as much higher accuracy and efficiency compared to other models.

The implementation of's core NLU algorithms on Xilinx FPGA boards enables the Messaging Classification Appliance to handle volumes of emails that would overwhelm standard, non-accelerated classifiers. As the classifiers auto-configurate using automated machine learning, there is no need for data science or AI experts to use the appliance effectively.

What makes's Messaging Classification approach different?

  • Fast processing inherent to’s NLU approach comes close to real-time processing when combined with Xilinx accelerating hardware
  • Understands the meaning in messages, even if they are short, and classifies them accurately, even if they do not use the same vocabulary as the filters
  • Meaning-based algorithm drastically reduces false positive rate

Which companies can benefit from's Messaging Classification approach?

  • Companies having to comply to strict regulatory standards in respect of their communication
  • Companies processing a high volume of emails on a daily basis
  • Companies monitoring customers’ intents on social media

How an International Transportation Company Optimized High Volume Email Processing in Customer Centers

view case study

About the Messaging Classification Appliance

  • Pre-packaged software and hardware solution
  • Combines Xilinx FPGA-based Alveo accelerator cards with AI-based classification algorithms
  • Can filter, classify, and route streams of messages in real time at massive scale
  • Available in Q1 2020
read the press release