Date of Award
5-30-2024
Document Type
Dissertation
Degree Name
Doctor of Philosophy (PhD)
Department
Information Science
First Advisor
Xiaowei Xu
Abstract
In the burgeoning fields of artificial intelligence (AI) and natural language processing (NLP), Large Language Models (LLMs) have emerged as powerful tools for understanding complex textual data. This dissertation focuses on the novel customization of LLMs for enhancing causal inference in pharmacovigilance and improving entity matching for data quality—two critical challenges in healthcare analytics and data management. Through an in-depth exploration of encoder and decoder LLMs, this study illustrates how domain-specific customization can significantly advance the processing and interpretation of textual information. For pharmacovigilance, it demonstrates how tailored LLMs can extract causal relationships from adverse event reports, offering a new paradigm in drug safety monitoring. In data quality, the research shows the potential of customized LLMs to accurately link records across datasets, a key process in data integration efforts. Empirical evaluations reveal that these customized LLMs outperform traditional approaches, providing significant improvements in both causal inference and entity matching tasks. This dissertation contributes to the AI and NLP literature by presenting a framework that leverages the strengths of both deterministic and probabilistic methods, addressing the evolving challenges of data quality and integration. The findings not only highlight the effectiveness of customized LLMs in specific applications but also suggest broader implications for employing AI-driven solutions in healthcare and data management.
Recommended Citation
Wang, Xingqiao, "Customization of Large Language Models for Causal Inference and Data Quality" (2024). Theses and Dissertations. 1194.
https://research.ualr.edu/etd/1194
