An Empirical Evaluation of Active Learning Strategies in Predictive Coding

Legaltech 2018

Please join Navigant’s Managing Directors Nathaniel Huber- Fliflet and Jianping Zhang at Legaltech 2018 for their session on “An Empirical Evaluation of Active Learning Strategies in Predictive Coding.” This session will take place on Wednesday, January 31st, from 1-2 PM, and is CLE eligible.

Active learning has drawn the attention of the legal community because it offers the potential to make the predictive coding process even more effective. But it turns out that different active learning strategies result in different levels of effectiveness. Following up on 2017’s Legaltech session: “Predictive Coding: Deconstructing the Secret Sauce,” this group of collaborators performed a new study related to Active Learning. This study demystifies active learning and examines popular active learning strategies to understand their effectiveness and practical application. This session will demonstrate that popular active learning strategies perform differently across data sets and the optimum performance of a given strategy may be achieved in the middle of the iterative process.

This session will highlight:

• The technical process and methodologies behind active learning

• The strengths and weaknesses of popular active learning strategies

• The experimental results in evaluating different active learning strategies

• The insight of active learning in predictive coding


Other featured speakers include:

• Rishi Chhatwal, AT&T

• Robert Keeling, Sidley Austin



New York Hilton Midtown  

1335 6th Ave.

New York, NY 10019

Back to top