In the past year we have seen more than a fair amount of discussion in e-discovery conferences and the surrounding blogosphere about the subject of predictive coding, also known as technology-assisted or computer-assisted review. In the wake of the landmark decision in Da Silva Moore v. Publicis Groupe, 287 F.R.D. 182 (S.D.N.Y.), and subsequent precedent, predictive coding is being considered by a growing portion of the legal community as a reasonable, more efficient alternative in appropriate cases to linear, manual review, or sole reliance on keyword searches. However, lawyers have an opportunity to do more than simply take advantage of the Da Silva Moore case for being smart in e-discovery; arguably, the decision also paves the way for a broader discussion with our clients on how lawyers can provide value based on the use of analytics across a wide variety of legal settings in the emerging world of big data.
First, a word about predictive coding. A good definition for our purposes here is that “predictive coding is the process of using a smaller set of manually reviewed and coded documents as examples to build a computer-generated mathematical model that is then used to predict the coding on a larger set of documents. It is a specialized application of a class of techniques referred to as supervised machine learning in computer science,” as Rajiv Maheshwari wrote in “Predictive Coding Guru’s Guide.”
