Digital White Papers

ILTA WHITE PAPER: JUNE 2014 WWW.ILTANET.ORG 38 PREDICTIVE CODING: REVOLUTIONIZING REVIEW OR STILL GAINING MOMENTUM? In the legal sector, e-discovery has finally been taken by storm with an advanced technology that other industries have employed for many years — predictive analytics and coding, which employ standard culling techniques such as date filtering and deduplication. Predictive analytics are also powered by algorithms that incorporate machine learning and computer-assisted review. With this technology, relevant case concepts and documents can be identified earlier in the discovery process, minimizing the time attorneys spend reviewing documents. To examine whether predictive coding has changed the review process or is still in its infancy, we spoke with Jessica Ross, a litigation attorney at Latham & Watkins LLP, and Jeanette Weir, the Senior Litigation Support Project Manager at WilmerHale, LLC. They provided insights on a wide range of issues around predictive analytics and coding, including what drives the decision to use this technology, how case law has influenced its use, the most common concerns around employing predictive coding and analytics during discovery, demystifying tools and processes, and how predictive coding will evolve. How do you define predictive coding? Jessica Ross: I define predictive coding as a technology-assisted categorization of a document population based on attorney input and verification. More important, predictive coding is a process that relies heavily on attorney decision- making to guide sophisticated software to find relevant electronic data. Jeanette Weir: Predictive coding is a combination of technology and human review to identify key and responsive documents. Predictive coding and analysis do not work on their own; they are part of a plan that has elements of human expert review and machine learning of like documents, with both working together to classify documents for production. Oftentimes people assume that with predictive coding you will produce or withhold documents that no human has looked at. That is not necessarily the case. Predictive coding can empower any review. How has predictive coding affected your firm and your current role? Jessica: Predictive coding has enhanced my role at the firm because it has added a new option to our toolbox. As an e-discovery specialist, I serve as a resource to our case teams to provide guidance in choosing a workflow and tools best suited to the case at hand. Predictive coding is now one of many options we have at our disposal to craft an efficient Jessica Ross Jessica Ross is an e-discovery attorney in the Latham & Watkins New York office. She works primarily on developing and implementing best practices related to electronically stored information. Jessica also develops training programs on electronic discovery and regularly consults with clients and firm attorneys on electronic discovery issues including preservation, collection, review and production of documents, discovery sanctions, privilege and the efficient use of technological resources in litigation. Contact her at jessica.ross@lw.com. Jeanette Weir Jeanette Weir is the Litigation Support Project Coordinator at WilmerHale, with nearly 15 years of experience in legal services. She has a wide array of expertise that is continually expanding through her current role and prior positions with Debevoise & Plimpton, Squire Sanders and Linklaters. Jeanette has a strong ability to assess business objectives and challenges, define and prioritize strategies and guide project scope from concept to completion in electronic discovery for complex litigation/controversy matters. Contact her at jeanette.weir@wilmerhale.com.

• Cover