Deep learning for third-party risk identification and evaluation at Dow Jones

Description

For more than 16 years, Dow Jones has supplied risk and compliance data to banking and financial institutions, corporations, and governments across the world, sharing defined, structured content sets of people and entities used to manage third-party risk: anti-money laundering, antibribery and corruption, sanctions, and reputational risk. In order to achieve a comprehensive coverage guided by international regulation and guidance since 2002, the company follows very high editorial standards and research methodologies, combined with state-of the-art machine learning techniques, to manage 30 risk categories 24 hours a day in over 70 languages.

Dow Jones recently wanted to apply a new approach to the existing content delivery pipeline with the objectives to eliminate low-level, repeatable, manual processes, enabling researchers to focus on strategic tasks; gain intelligence from global media and research tools, scanning and monitoring almost 2 million articles per week; and achieve near-real-time risk data detection and delivery capabilities. Yulia Zvyagelskaya and Victor Llorent explain how the company created an AI-powered Risk & Compliance data research solution that uses natural language processing for risk profiles creation and management. Along the way, they highlight the unstructured data preprocessing stage, model selection criteria, and neural networks parameter tuning processes to provide scalability and performance in order to achieve mentioned key objectives.