Tim Carmody: CTO at IPC, a Leading Financial Trading and Communications Tech Firm, Explains Role of Natural Language Processing in Finance

We recently caught up with Tim Carmody, Chief Technology Officer of IPC, a global leader in financial trading and communications technology solutions. Carmody, who has several decades of experience in designing and leading complex technology solutions for the trading community, discussed how Wall Street companies are using natural language processing.

Carmody, a graduate in Systems Engineering from Boston University, also talked about what NLP will look like in a decade from now.

What is Natural Language Processing, and how is Wall Street using it?

Tim Carmody: “Natural Language Processing (NLP) is the ability of a computer program to recognize and understand human conversations as they’re spoken. NLP is a technology that has grown by leaps and bounds over the past decade due to advances in machine learning, which allows computers to understand and process human speech and text patterns to a far greater degree than previously possible.

NLP has enormous applications on Wall Street in several different areas. For example, with regard to compliance, in many instances, traders are negotiating and executing trades by voice. Today, those trades are being recorded and then separately transcribed for posterity and keyword searches as needed — that is an extremely arduous process – but NLP can help solve these issues.

Bear in mind these traders are speaking amid very busy and loud trading floors and in Wall Street jargon. NLP can decipher this complexity quickly and operate in real-time to capture as searchable and actionable data. However, this requires real-time access to pristine audio and very sophisticated, trained NLP programs.

There are many other facets to NLP on Wall Street — certainly, traders appreciate the ability to have high value and significant trades captured and transacted accurately by merely speaking them. And for surveillance, NLP provides great value around analytics capabilities by turning voice data into actionable insights.

For example, through the voice data it captures, NLP can better illuminate trader-customer interactions by answering questions like: Which customers generate the most business? Which customers request prices but don’t trade? What percentage of price requests lead to a booked trade?”

What types of customers turn to IPC for NLP? What can IPC do in this field?

Tim Carmody: “A wide variety of customers on both the sell and buy sides come to us for NLP solutions, as there are several different areas (e.g. execution, analytics, compliance) in which NLP is extremely useful for financial traders and institutions. Our flagship NLP product called ‘Blotter’ pairs transcription and NLP for an array of needs by some of the largest financial institutions in the world. For instance, Blotter leverages an NLP engine to convert over-the-counter (OTC) voice quotes into a structured data feed.

It has been very illustrative for us at IPC to see the many ways in which NLP unlocks hidden value; for example, with Blotter, trader voice and chat conversations can be converted into a price data feed to be sold to users’ customers for an additional source of revenue.

One area with regard to compliance I would also note is that regulations like MiFID RTS 27 require market participants to maintain records of all trades, but this information is often lost; Blotter can extract each quote and trade from voice communications and pass that to the compliance database.”

If someone starting their career wanted to work with NLP in finance, what advice would you give them?

Tim Carmody: “That’s a great question — it depends in what respect you’d like to work with NLP in finance, because, on the trading side, it’s quickly becoming ingrained into everyday processes; we’re building it so that, by design, you don’t need any special tool kit to use NLP as a trader.

Now, if you want to build the underlying NLP, that’s definitely a job in which coding comes into play. If you’re already familiar with the basics of coding, there’s some great open-source NLP packages for Python out there like Natural Language Toolkit (NLTK), and spaCy.”

Where do you see NLP in 10 years?

Tim Carmody: “Well, if you had asked me this question 10 years earlier, I don’t know if I could’ve predicted how far we’ve already come in NLP, so I can’t imagine what the future will hold. I would say at a very high level that, based on our current progress, it seems likely that soon we’ll find speaking to a computer and speaking to a person will become indistinguishable. The potential impact for Wall Street and beyond is profound.”

Sponsored Links by DQ Promote



Send this to a friend