Legal Natural Language Processing (DSI-SRP)
This DSI-SRP fellowship funded Henry Savich to work in the laboratory of Professor Brian Fitzpatrick in the Vanderbilt Law School during the summer of 2020. Henry graduated in May 2021 with a major in Mathematics and minors in Physics and Scientific Computing.
The Legal Natural Language Processing (NLP) project, funded by this fellowship, aimed to use state-of-the-art machine learning methods to read and understand U.S. legal documents. The project is at the intersection of law, data science, and linguistics. The primary goal is to create artificial intelligence that is capable of reading legal documents about a lawsuit and understanding important information about the case, such as what the case was about, how the case was resolved, who the parties were, what legal tests the court used, etc. By extracting these variables and outputting them in a convenient form, AI can make it possible to analyze much larger sets of cases than can be analyzed now with traditional methods.
The project is currently testing documents about class action lawsuits. Henry was instrumental in refining the machines to read and understand these documents. He attended a number of workshops, which enabled him to effectively collaborate on the project through versioning with git, sharing with GitHub, Python, and machine learning. He also learned to use advanced software such as nbdev, which streamlines collaboration and enables fast sharing of work to the larger Python community. After the SRP, he continued to work on the project the following semester with 4 new participants, and was the technical lead in teaching the other students to use the tools which he had learned during SRP. Henry provided helpful feedback on coding through pull requests.
The results on this project have thus far been limited, but we are optimistic they will improve as new machines become available. The work will continue for the foreseeable future.
In addition to receiving support through a DSI-SRP fellowship, this project was supported and facilitated by the DSI Data Science Team through their regular summer workshops and demo sessions.