From Data to Insights - How FINRA Uses Machine Learning

FINRA receives over 60,000 triggering events annually. The events range from investor complaints, regulatory tips, arbitration claims, and member firm disclosures. During very high market volatility periods, FINRA receives spike volumes of triggering events and all of these events have to be triaged within tight SLAs.

This presentation will focus on how FINRA experimented with multiple machine learning models to predict the severity and allegations that are present in the incoming triggering events to FINRA. Furthermore, the presentation will focus on a serverless architecture to build a near real-time solution that helps FINRA investigators to focus on high-priority events, thereby safeguarding market integrity and protecting investors from bad actors. 

Speakers: 

Faarah Dordy, FINRA

Faarah is an accomplished technology leader with over 15 years of experience in data solutions and product management. At FINRA she has been responsible for designing and implementing technology solutions aimed at detecting behaviors of regulatory interest in the U.S. Capital markets. She leads a team of data analysts and product managers who build advanced algorithms and products to extract meaningful insights from the massive amounts of regulatory events and market transactions collected by FINRA. She lives with her husband and two daughters in Southern California and loves to explore different interesting eateries in the area.

Aja Klevs, FINRA

Aja is a data scientist at FINRA with over 5 years of experience in machine learning. She has a rigorous technical background with a bachelor’s in mathematics from UC Berkeley and a master’s in data science from NYU. During her two-and-a-half years at FINRA, she has spearheaded many machine learning projects from quantifying broker risk to automating audit reviews. In addition to math, Aja has a passion for art, contemporary fiction, and American history.

Hari Narayanan, FINRA

Hari is an Engineering Manager with over 25 years of experience in the IT industry. He is working to leverage data to solve complex business problems in the area of Insider Trading, Financial crimes. Recently his focus has been on the build of a regulatory workspace that provides task management, workflow, and matter management capabilities for regulatory investigators and examiners.