It’s no secret that the flood of digital communications from email, messaging, social networking and other sources exasperates our daily lives. Enterprises, in particular, are overwhelmed with communications from inside and outside their organizations, with content that needs to be acted on in a timely manner. And today’s Covid-19 environment of resource constraints, social distancing, and working from home only compounds the problem.
Consider these facts:
In 2019, more than 128.8 billion business emails were sent and received per day.
The average office worker receives 120 emails every day and spends a staggering 2.6 hours each day reading and answering them.
Often, an incoming email doesn’t even reach the right person within the organization quickly. This impairs productivity in matters where speed is critical, resulting in operational delays and rising costs.
AI-based Natural Language Understanding (NLU) solutions have the potential to deal with the deluge of email by understanding the content of the messages -- automatically and accurately classifying them, and then taking appropriate action such as routing, tagging, or handing off the message to an application. However, the sheer volume of messages represents a real challenge to the computing capacities of most existing solutions which, also, lack the accuracy of human workers to understand the meaning of ambiguous messages.
Enter the Xilinx powered Cortical.io Message Intelligence. Cortical.io, a leader in AI-based NLU, has introduced a new supercomputing appliance that combines the company’s proven Semantic Folding method with Xilinx’s Alveo® U50 data center accelerator cards to deliver a purpose-built appliance that filters, classifies, and routes streams of messages in real-time at a massive scale. The out-of-the-box solution is now available and can be easily installed and configured. The appliance provides human-level accuracy in filtering and classifying messages, while enabling businesses to quickly process and respond to messages, of any kind, while mitigating legal risks. Users can create custom semantic classifiers and filters with minimal effort. Cortical.io solutions are based on neuroscience and can be quickly trained without supervision in the specialized vocabulary of any business domain and can function across multiple languages.
Semantic Folding-based NLU algorithms such as search, classify, or extract is implemented as binary operations on semantic fingerprints. These algorithms, implemented in hardware, take advantage of the parallelism possible with our Alveo accelerator cards to provide an order-of-magnitude performance increase over a standard CPU-based implementation.
With the Message Intelligence, the first release of Cortical.io's Semantic Supercomputing Appliance, Cortical.io harnesses the adaptive compute acceleration of our Alveo cards to perform several complex tasks in a highly parallel manner. The system uses Cortical.io's unique algorithm to create a semantic meaning of an entire text, for example, a message to be classified. This allows for a very efficient classification of the message, similar to how a human would "understand" the content of the text. In our example, the words in the database are divided into "frequent" (most frequently occurring in texts) and infrequent ones. The frequent terms are matched in extremely fast, using the Alveo U50 card’s massively parallel on-chip memory, and the infrequent terms are indexed in parallel on the chip and retrieved from the U50 card’s 8GB of High-Bandwidth-Memory (HBM2). The diagram below shows the parallel indexing portion of the U50 card’s FPGA design. The massively parallel frequent term matching part is not shown as the number of parallel elements would exceed a useful graphic resolution.
Message Intelligence is the first in a series of Cortical.io and Xilinx-powered Semantic Supercomputing appliances. The appliance comes as a pre-packaged software and hardware solution with pre-built filters and classifiers that can be immediately applied to high volume message streams in an enterprise environment.
Together with Cortical.io we’re delivering Semantic Supercomputing appliances to enable the broad implementation and deployment of AI solutions for automating business processes and solving the most challenging use cases requiring human understanding, decision making, and execution.