Attorney Brown Teaches Law School Course on eDiscovery Technologies

Attorney and Adjunct Professor Shannon Brown taught a course on eDiscovery technologies such as keyword search, technology assisted review (TAR), predictive analytics, and predictive coding.

Unlike typical eDiscovery courses, which often gloss-over the technologies, law students became deeply engaged in the technical aspects of eDiscovery including completing a hands-on, simulated eDiscovery project using a freely available, predictive coding, reference platform and parts of the Enron dataset.

Students also learned in the law school course about the technical progression in eDiscovery from early TIFF load files, to keyword search, to technology assisted review (TAR) to predictive coding. Importantly, the law school course introduced the students to how some of the primary algorithms work–including keyword search indexing, Support Vector Machines (SVMs), logistic regression, clustering technologies (basic k-Means and k-Nearest-Neighbors), and Bayesian decision systems.

Students learned about several metrics used to evaluate eDiscovery software performance such as accuracy, recall, precision, F-scores, and probabilities.

A significant part of the law school course on eDiscovery also addressed predicate preprocessing issues, which have legal consequences, such as de-NISTing, understanding the difference between forensics and eDiscovery, feature selection, data matrices, stemming, natural language processing, latent semantic indexing, and generalization (overfitting and underfitting).

At the conclusion of the law school course, students were expected to be able to identify the types of technologies used in TAR, predictive coding, keyword, or other tools and be able to apply them to eDiscovery technical problems.

Other class topics included eDiscovery project management, understanding Big Data, basic case law, and eDiscovery best practices.

Widener School of Law-Harrisburg held the 16-hour, intensive program. The program is believed to be the first of its kind in the country–a course intensively focusing on understanding and applying the technologies and a notable simulated case using predictive coding.