Menu
Matthew A. Lanham
Matthew A. Lanham

Matthew A. Lanham, CAP-X

Assistant Professor of Business Technology and Analytics
Butler University’s Lacy School of Business
Bio

Matthew A. Lanham is an Assistant Professor of Business Technology and Analytics at Butler University’s Lacy School of Business. He earned his Ph.D. in Business Information Technology from Virginia Tech and researches the interface of predictive and prescriptive analytics, with a focus on data analytics for good. He serves as VP/President-Elect of the INFORMS Analytics Society and is an elected member of the INFORMS Analytics Certification Board and the SAS Faculty Advisory Board. Lanham received the 2025 inaugural INFORMS Data Mining Society Teaching Award and was named a 2025 Poets&Quants Best Undergraduate Business Professor for classroom impact and global leadership of the National Data4Good Analytics Competition. His applied AI work spans partnerships with organizations including SIL Global, Prediction Guard, Puef.ai, ZETEC.ai, Marion County Courts, and the Indiana Supreme Court.

 


Retrieval-Augmented Decision Support for the Judiciary: A Domain-Grounded, Explainable Analytics System for Legal Review

Judges in large trial courts routinely face hundreds of pages of transcripts and exhibits per matter, making thorough, consistent review difficult under tight time constraints. This presentation introduces a secure, retrieval-augmented decision-support system that transforms unstructured court records into a judge-ready interface for fast semantic search, citation-backed Q&A, and structured summarization. The workflow integrates descriptive analytics to convert transcripts into speaker-labeled, metadata-rich segments; predictive analytics to build legal-aware embeddings for high-fidelity retrieval; and prescriptive, task-oriented outputs that directly support preparation and prioritization decisions. Our core contribution is a domain-grounded RAG architecture purpose-built for judicial records, combining legal-focused ingestion rules, context-preserving chunking, and citation-constrained generation to reduce hallucination risk and improve factual stability on large, sensitive case files. In pilot use, the system reduced transcript review from roughly 3–4 hours to about 20 minutes while improving user confidence via direct links to supporting passages. The approach has been validated through demonstrations and feedback from 15 Marion County judges and is being rolled out to additional U.S. court jurisdictions, offering a practical model for responsible GenAI adoption in high-stakes public-sector workflows.

Essential / Professional / Leadership
Back To Top

The Institute for Operations Research and the Management Sciences

5521 Research Park Drive, Suite 200
Catonsville, MD 21228 USA

phone 1 443-757-3500

phone 2 800-4INFORMS (800-446-3676)

fax 443-757-3515

email

Get the Latest Updates

Copyright INFORMS. All Rights Reserved