PhD Dissertation Proposal by WANG Sha | Peer Group Analysis using Knowledge Graphs and Large Language Models for Business Optimization

ABOUT THE TALK

This dissertation explores advanced methods for peer group generation by integrating Knowledge Graph (KG) ontologies with Large Language Models (LLMs) to enhance business processes such as due deligence, price benchmarking, and performance evaluation. The research addresses two key components: first, the development of an OLAP-like tool that facilitates the exploration of unstructured text data (e.g., financial reports, news articles) through KG-based rollup and drilldown capabilities for peer group analysis. Second, it tackles the limitations of manually maintained KG ontologies by using LLMs to dynamically update and refine these ontologies, improving schema matching and keeping pace with rapidly changing data. This work aims to optimize peer analysis workflows, providing more efficient and accurate insights for real-world business decision-making.

ABOUT THE SPEAKER

WANG Sha is a PhD Candidate in Computer Science at the SMU School of Computing and Information Systems, supervised by Assistant Prof. LI Yuchen.  Her research is focused on unstructured data mining with KG and LLM.