For E-discovery Professionals, the Difference Between TAR, CAL and SAL Is Often Priorities

Computer at bottom with funnel catching cloud, graphs and other data types, with gears

Lilith Bat-Leah, director of data science and AI strategy at Driven, Inc. and Dr. Jermey Pickens, principal data scientist at EDRM Partner, OpenText, and both EDRM leaders of the Analytics and Machine Learning project, presented on the differences between technology assisted review (TAR), continuous active learning (CAL) and simple active learning (SAL) at ILTACON 2021.

Frank Ready, reporter on the tech desk at ALM Media, dove deep on the analytics panel for Legaltech News.

The panel, titled “Wait, Aren’t They the Same Thing?! The Technology Assisted Review (TAR) v. Continuous Active Learning (CAL) Duel,” featured Lilith and Jeremy discussing factors that could impact which flavor of analytics would reduce cost, risk and/or time.

CAL is often referred to as “TAR 2.0” and uses infinite, active learning based on relevance to review documents. SAL, on the other hand, is typically thought of as “TAR 1.0” and leverages finite, passive learning rooted in random or judgmental sampling. 

Dr. Jeremy Pickens, principal data scientist at OpenText

The panelists spoke about not only the math involved, and also take into consideration inferences about past projects and intuitions about what challenges the case and its data will present.

It’s worth articulating your assumptions just so you can stop and think for each assumption ‘am I comfortable defending this assumption in court?

Lilith Bat-Leah, director of data science and AI strategy at Driven Inc. 

Read Frank Ready’s full analysis here.

Author

  • Mary Mack is the CEO and Chief Legal Technologist for EDRM. Mary was the co-editor of the Thomson Reuters West Treatise, eDiscovery for Corporate Counsel for 10 years and the co-author of A Process of Illumination: the Practical Guide to Electronic Discovery. She holds the CISSP among her certifications.

    View all posts

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.