Bringing AI to Network Data

Corvil has been focusing on statistical analytics since it was founded, so machine learning and predictive analytics is a natural next step.

The growing interest in AI, machine learning and data science is something we have followed with great interest in Corvil. After all, we were founded in 2000 by researchers, including myself, who were pioneers in the use of advanced modeling and statistical analytics techniques long before the concept of data science even existed.

As AI evolved into a separate discipline – and data scientists became highly sought after resources – we set up a dedicated machine learning/data science unit that acts as a repository for the advanced mathematical techniques we apply to our products. We have grown and matured with the science, and made it our business to leverage increasingly sophisticated techniques for the benefit of our customers.

Augmenting People

There has been a lot of speculation about how AI and machine learning will replace the need for humans, but when it comes to network analytics the reality is a lot more dull than the ‘robots are coming’ scare stories might have us believe. The sheer volume of data makes it impossible for human beings to keep an eye on all of it, so yes, there is a valuable role for AI. But will people be replaced? Not for the foreseeable future.

For algorithms to produce meaningful results they need people to apply them to the areas that matter. Point them at data blindly and ask them to find things out on their own and they will come back with a huge number of anomalies, most of which won’t be important. Change is constant in a modern digital business; what might appear to AI as an anomaly might just be normal in fast-changing IT environments.

So we’re still a long way away from switching on machines that will monitor the network, automate everything, and solve all your problems. There are, however, a number of use cases where AI and machine learning already make a substantial difference.

Corrective Action

For IT operations teams, AI become an invaluable assistant in troubleshooting. By sifting through network data, it can identify the starting point and the wider context of an event, helping resolve anomalies more quickly while providing insights to make sure they never happen again.

We’ve delivered sector-specific use cases as well. For example, MIFID II mandates that financial traders have to stay within designated ratio limits, but it can be difficult in the course of a busy day to know how close they are to hitting the limit. Based on past experience, machine learning can predict where those ratios are likely to finish up at and alerts can be triggered to prompt traders to take corrective action.

What makes both of these scenarios possible – and good use cases for machine learning – is the data provided by Corvil machines. It is widely recognized in data science that the quality of the data is much more important than the algorithms used to model it. A good source of quality data – and lots of it – has been our differentiator in the past and will continue to give us an edge in the future as machine learning and predictive analytics become more prevalent.

The traditional challenge with network data was that the volume and diversity of data made it hard to separate meaningful events from noise. Encoded in lots of different ways, depending on protocols and applications, it can be opaque and difficult to understand. This is a problem we cracked many years ago. Bringing AI and machine learning to that process makes things even more interesting.

Corvil Introduces AI-Powered “Intelligence Hub”

Since the key to analyzing and interpreting data at scale is in using quantitative techniques, the next logical step for us is making AI, machine learning and data science more accessible to business, IT and security users. This is a problem we’re tackling with Corvil Intelligence Hub.

Our goal is to enable individuals with little to no data science expertise to leverage the power of AI for greater, more proactive actionable insight. As more teams become familiar with the types of intelligence AI can deliver when applied to this high-quality data, they will benefit from the ability to make better, faster decisions; improve operations; and enhance customer experience.

Learn more about Corvil Intelligence Hub
Explore the Solution   Watch Now
Download the Solution Brochure
Learn about Self-Service Intelligence for Electronic Trading

Fergal Toomey

Fergal Toomey, Chief Scientist
Pico is a leading provider of technology services for the financial markets community. Pico provides a best-in-class portfolio of innovative, transparent, low-latency markets solutions coupled with an agile and expert service delivery model. Instant access to financial markets is provided via PicoNet™, a globally comprehensive network platform instrumented natively with Corvil analytics and telemetry. Clients choose Pico when they want the freedom to move fast and create an operational edge in the fast-paced world of financial markets.