BIMSA >
BIMSA Digital Economy Lab Seminar
Textual Big Data and LLMs: Applications and Credibility Challenges in Economics and Finances
Textual Big Data and LLMs: Applications and Credibility Challenges in Economics and Finances
Organizers
Speaker
Time
Friday, April 11, 2025 3:00 PM - 4:00 PM
Venue
A3-2a-302
Online
Zoom 637 734 0280
(BIMSA)
Abstract
With the rapid development of the internet and computing technologies, textual big data such as annual reports of listed companies, analyst research reports, and social media data—has provided rich and efficient data sources for economic and financial research. Using text analysis techniques, researchers can extract important information from these unstructured datasets. This talk reviews the frameworks of traditional text data analysis and introduces recent applications of large language models (LLMs) in the economics and finance fields. Additionally, we discuss the limitations of LLMs in text analysis and the credibility issues they pose. Key challenges, such as the reproducibility of output results and insufficient model transparency, are examined, along with potential approaches to address these limitations.
Speaker Intro
Junda Wu is a PhD student at BIMSA and UCAS. His research interests include digital economy, applications of artificial intelligence in economics, and cross-market risk spillovers.