Migrated from eDJGroupInc.com. Author: Greg Buckles. Published: 2015-07-12 20:00:00Format, images and links may no longer function correctly.
The ‘TAR Wars’ have argued incessantly over the different approaches to training the analytic engine to extract the 1-3% of relevant items from your raw collections. Proponents of true random sampling (Herb Roitblat of Orcatec) have criticized ‘Active’ sampling based on user provided seed documents because the active approach might miss unanticipated concepts or documents. In other words, some approaches are great at finding what you already know about, while others might find relevant documents that you never even considered and eliminate potential reviewer bias. Given the selective recall of many corporate custodians, diligent counsel will want to know the good and the bad before trying to settle a matter. Savvy counsel, providers and consultants have used multiple approaches in the primary review and QC stages to optimize recall while tackling ever larger collections. The conceptual navigation approach starts with a visualization or inventory of conceptual clusters to let the collection ‘speak’ for itself. Brainspace, Tunnelvision and other platforms have advocated this top down conceptual navigation to know what you have before you start training your relevance model. Catalyst has coined the term ‘Contextual Diversity Sampling’ to describe their system to find topics that reviewers might have missed in the Continuous Active Learning process. There are all kinds of jokes about how diversity can make us all PC. Call it what you want, I am pleased to see more mature, practical analytic workflows hitting the market. We all know that manual linear review of 100% of collections is neither accurate nor practical in the face of the corporate data hoarding juggernauts. So we need review solutions that even conservative counsel will get behind to tame skyrocketing discovery budgets. Keep asking hard questions and sending me feedback on both hype and success.
Greg Buckles wants your feedback, questions or project inquiries at Greg@eDJGroupInc.com. Contact him directly for a ‘Good Karma’ call. His active research topics include analytics, mobile device discovery, the discovery impact of the cloud, Microsoft’s Office 365/2013 eDiscovery Center and multi-matter discovery. Recent consulting engagements include managing preservation during enterprise migrations, legacy tape eliminations, retention enablement and many more.
Blog perspectives are personal opinions and should not be interpreted as a professional judgment. eDJ consultants are not journalists and perspectives are based on public information. Blog content is neither approved nor reviewed by any providers prior to being posted. Do you want to share your own perspective? eDJ Group is looking for practical, professional informative perspectives free of marketing fluff, hidden agendas or personal/product bias. Outside blogs will clearly indicate the author, company and any relevant affiliations.