Benign Ihugba Master’s Thesis Defense, Monday, November 24, 2025 @ 1:30 pm Central Time
November 24 @ 1:30 pm - 2:30 pm
COMMITTEE CHAIR: Dr. Lin Li
TITLE: KNOWLEDGE ACQUISITION ON MASS-SHOOTING EVENTS VIA LLMS FOR AI-DRIVEN JUSTICE
ABSTRACT: Mass-shooting event poses a significant challenge to public safety which generates large volumes of unstructured textual data that makes it difficult to conduct effective investigations and formulation of public policy. The proposed knowledge acquisition system utilizes Large Language Models (LLMs) with few-shot prompting to extract vital information from news articles and police reports and social media content at high speed. The system uses entity recognition to detect vital information which includes offenders and victims and locations and criminal instruments that support legal investigations. Experimental tests conducted on actual mass-shooting datasets show GPT-4o performs best for mass-shooting NER because it produces the highest scores in Micro Precision and Micro Recall and Micro F1-scores. The o1-mini delivers competitive performance while using less resources more efficiently for basic NER applications. It is also observed that when increasing the number of shots, it enhances the performance of all models, but the gains are more substantial for GPT-4o and o1-mini, highlighting their superior adaptability to few-shot learning scenarios.
Room Location: S.R. Collins Building, Room 111L
