Following a New York federal court’s approval of predictive coding in the Rio Tinto case, attorneys are increasingly using predictive coding to expedite discovery when faced with large quantities of documents.

While predictive coding can save significant time and money in discovery, parties often disagree on the methods and protocols for implementing it, potentially negating the time and cost savings. Courts or the involved parties may want to push for the appointment of an eDiscovery Special Master as early in a case’s lifecycle as possible to mediate protocol disputes and provide guidance on technical issues, hopefully recapturing the benefits of predictive coding and saving the parties time and money.

What is Predictive Coding?

Predictive coding is, essentially, the use of keyword search, filtering and sampling to “teach” a computer to identify responsive or privileged documents. While there are multiple variants, the basic process is as follows. First, a set of documents, referred to as the “Seed Set” is pulled from the entire corpus of documents for review. A subject matter expert, intimately familiar with the case, then codes the Seed Set for responsiveness and privilege. A computer analyzes the expert’s coding, attempts to learn what makes documents responsive or privileged, and codes a new set of documents. The expert then corrects the computer’s mistakes, and the computer is tasked with coding another set of documents, which are then reviewed by the subject matter expert. This process is done iteratively until the computer can hit certain statistical benchmarks. Once the computer is properly trained, it codes the entire document corpus.

To read the full article, go to Legal Executive Institute.