Professor Vern R. Walker co-authored a paper entitled “Introducing LUIMA: An Experiment in Legal Conceptual Retrieval of Vaccine Injury Decisions using a UIMA Type System and Tools” that was peer-reviewed and accepted for presentation at the 15th International Conference on Artificial Intelligence & Law (ICAIL 2015), held in San Diego, June 8-12, 2015.
The paper’s co-authors were Matthias Grabmair and Kevin D. Ashley of the Intelligent Systems Program at the University of Pittsburgh, as well as Eric Nyberg, Ran Chen, Preethi Sureshkumar and Chen Wang of the Language Technologies Institute of Carnegie Mellon University.
The paper was one of a series resulting from a collaborative project on automated argumentation mining between Hofstra Law’s Research Laboratory for Law, Logic & Technology (LLT Lab), which Professor Walker directs, the University of Pittsburgh and Carnegie Mellon University.
This paper presents some first results from a proof-of-feasibility experiment in conceptual legal document retrieval in a particular domain (involving vaccine-injury compensation).
The conceptual markup of documents is done automatically using LUIMA, a law-specific semantic extraction toolbox based on the Apache UIMA framework (Unstructured Information Management Architecture). UIMA provides the architecture for creating a pipeline of software analytics that can automatically annotate a text for its meaningful roles (for example, inference roles in argumentation or reasoning).
The LUIMA system developed to date consists of modules for machine-learning-based annotation of entire sentences, for automatic annotation of sub-sentence elements, for basic document retrieval using Apache Lucene, and for machine-learning-based re-ranking of retrieved documents.
This paper compared the resulting LUIMA rankings to baseline rankings created using a commercial legal information system (WestlawNextTM), with the results outperforming the commercial system for most tested queries.