Recommind, Deloitte to Organize Banks’ Data

June 24, 2015, 10:23 PM

Recommind announced on Wednesday that it has introduced a new tool, to be managed by Deloitte, that can organize and extract data from complex over-the-counter derivative legal agreements.

The new product, called the Perceptiv Derivatives Contract Analysis, utilizes the company’s machine learning technology to sift through the OTC agreements, identify key data points, such as termination points, bankruptcy contingencies and various others, and classify each agreement based on the language used.

Until now, most banks have been looking for this data “manually,” according to Mike Lines, who is heading the project for Recommind.

Lines argued that Recommind’s automated software would make the process of taking data out of OTC contracts more accurate and efficient for the banks as the Dodd-Frank Act, as well as other regulatory requirements in Europe, increases the pressure to make such data more readily available.

Deloitte will be managing the software and the relationship with the banks, Lines said, in a brief interview with Big Law Business. Below is a lightly edited transcript.

Big Law Business: Who is the target client for Perceptiv?

Lines: The big sell-side organizations, the big tier one banks. Basically, the problem the banks have is a shortfall of data, and the data they do have is sorted into different systems. And to be frank, the banks don’t really trust the data they have. On the operations side, they don’t want to spend millions of millions of pounds going to lawyers basically scraping data from these agreements. They don’t want to send White & Case or Clifford Chance 30,000 agreements and ask them to get [the data] off the agreements.

We’re [also] very focused on the largest tier one banks. You’re solving problems that cost the bank tens of millions of pounds per year. We’re also getting loads of people from the buy side. People coming to us saying we’ve got 300 or 400 agreements and we really need to know what’s in them. So we think it’s an important business problem to solve.

Big Law Business: What will Deloitte’s role be?

Lines: They’re really managing the service. They’ll be operating the software. They’re providing people to provide quality assurance to the banks. It’ll be a mixture of lawyers and smart people who understand the process.

Big Law Business: What’s the underlying technology and how does it work?

Lines: The contracts we’re dealing with here are PDFs, just flat data. They’re not searchable. The core product that Recommind has in the back end is used in all of our products. Basically, what we’ve built here is new workflows, and data models.

There’s an interaction between people and the technology to calibrate it. The system has underlying algorithms that it uses to continuously improve its understanding of similar language, similar phrases. [Perceptiv] is a service that we don’t customize for any single bank. From a machine learning point of view, we build up a much bigger library of phrases and language that we can use.

To read more articles log in.

Learn more about a Bloomberg Law subscription.