Human-Readable Description Extraction from Tabular Data

This project proposal aims to tackle a significant challenge in fintech: extracting human-readable descriptions from tabular financial data. Interpreting vast amounts of structured data in the dynamic financial technology landscape is crucial for informed decision-making and compliance reporting. The project seeks to develop a robust system by leveraging open-source Large Language Models (LLMs) tailored to Verafin’s needs. By systematically evaluating LLM performance, refining prompt engineering techniques, and benchmarking against industry standards, the project aims to optimize accuracy, efficiency, and scalability. Through this endeavor, Verafin enhances its data interpretation capabilities and reinforces its position as a leader in fintech innovation.

Faculty Supervisor:

Sarah Power

Student:

Partner:

NASDAQ Canada Inc

Discipline:

Computer science

Sector:

Finance and Insurance; Artificial Intelligence

University:

Memorial University of Newfoundland

Program: