As a relative newcomer, Big Data is the topic at the keynote session on “The Judicial Perspective – Managing Big Data, Proportionality, Data Security, and Privacy,” with U.S. Magistrate Judge Andrew Peck of the Southern District of New York, Senior Judge Michael Baylson, U.S. District Court for the Eastern District of Pennsylvania, and Circuit Judge Thomas Vanaskie, U.S. Court of Appeals for the Third Circuit.
Big Data obviously means big challenges in e-discovery, and vendors at LegalTech will be hawking their flavors of technology-assisted review, often called predictive coding, “that appl[ies] complex algorithms to large heterodox data sets to extract theoretically more meaningful information than traditional analytics.” Organizations are increasingly realizing the benefits of utilizing technology-assisted review and other advanced analytical tools to get a handle on Big Data to meet e-discovery objectives.
Technology-assisted review software analyzes decisions about relevance from senior lawyers on a sample of data. The software compares the lawyers’ coding against each document’s content, determining the criteria that make a document more likely to be relevant or privileged. Then the technology engine applies the lawyers’ decisions across the data collection. As reviewers feed more information to the system as case information evolves, the accuracy and defensibility of the process should increase: through the iterative use of statistical sampling and quality control techniques, the software refines its decision-making ability.
The technology-assisted review process yields a set of documents ranked for responsiveness that counsel can assign to the appropriate level of reviewers. Accelerating the review of high-priority documents and limiting the review of less-responsive documents can add up to significant cost savings while addressing some of the challenges of Big Data – at least at the most basic level, sheer scale that makes putting human eyes on every document nearly impossible.
However, Mark Beyer, research vice president at Gartner, cautions that “CIOs should recognize that the technology is still relatively young and will continue to evolve over the next few years; as a consequence, they should ‘develop a strategy for continuous upgrades and skills to match the speed of technology for the next six years.’”
Not All Technology-Assisted Review Solutions Are Created Equal
The key word is “skills.” Adding the knowledge of skilled experts to the process can amplify the accuracy and defensibility of the decisions made by a technology-assisted review solution. In fact, these experts are arguably just as vital to the process as the technology itself. For example, linguists can guide parties in the selection of the keywords likely to generate the most responsive results. In Why Keyword Search Won’t Go Away and Revisiting the Relationship Between Keyword Search and Technology-Assisted Review in Judge Scheindlin’s National Day Laborer Opinion, we highlighted recent cases where the court offered valuable advice for best practices in applying technology-assisted review with other methods such as search, while statisticians can apply their expertise to validate the reliability of sampling techniques and the quality of the results.
We hope the judicial panel at LegalTech acknowledges the value of this type of expertise, which is hard to come by in the e-discovery market. But if you look in the right place, you just might find it (Booth #1510 at LegalTech).
About the Author