FINRA receives hundreds of thousands of documents and big data each year from stockbrokers and investors to be reviewed and analyzed. They are looking for the information about people, places, and things mentioned in these documents. It is a labor-intensive process to extract entities from free text and semi-structured data, let alone connect those entities to each other and identify significant regulatory patterns.

In our talk, we will share our experience building a knowledge graph data pipeline that starts from ingesting massive amounts of data from varieties of data sources to leveraging graph data to identify patterns that provide business value.

Attendees will learn:

• How both structured and unstructured data come together
• Statistical and machine learning approaches for sanitizing data
• Automating data documentation
• How we use AWS services to provide data to other users

Speaker: Pragnya Gandhi

Senior Principal Architect
FINRA

Pragnya Gandhi has over three decades of experience in everything data. She specializes in architecture and designs of complex systems and enterprise data platforms. Today she supports multiple advanced analytics initiatives as a Senior Principal Architect at FINRA.

Speaker: Chen (Anna) Zhang

Chen (Anna) Zhang is a Senior Developer at FINRA. From front-end to back-end, from services to data, she has a wide range of technical skills. However her heart lands on data. She has been providing data engineering support in building the knowledge graph for over 6 years.