July marked the 30th anniversary for Legaltech West. Attorneys and in-house counsel gathered to hear technology assisted review (TAR) pioneers – John Tredennick, Maura Grossman, Gordon Cormack and Emi Ohira – lead a session titled, “Taking TAR to the Next Level: Recent Research and the Promise of Continuous Active Learning.”

TAR (in this context) is a process for prioritizing a collection of documents so the relevant ones rise to the top, thereby reducing time wasted reviewing non-responsive material. Using a computerized system that harnesses human judgments on a smaller set of documents, the technology extrapolates those judgments to the remaining document collection.

The speakers discussed three TAR methods: continuous active learning (CAL), simple passive learning (SPL) and simple active learning (SAL). It can be difficult to keep these straight, but that doesn't mean that some review processes aren't better than others. I prefer to lump them into two categories: CAL, which constantly revises the most likely relevant documents based on every piece of input from reviewers, versus the others. The others (SAL and SPL) can be very effective as any technology that can help you reduce the amount of non-responsive information you have to look at is positive, however CAL is likely to be the most effective. To determine which technology is best for your next case, let's take a closer look at a few commonly used tools – Catalyst's CAL, Eclipse, and Clearwell.