Contextual diversity refers to documents that are different from the ones already seen and judged by human reviewers. The contextual diversity algorithm identifies documents based on how significant and how different they are from the ones already seen and then selects training documents that are the most representative of those unseen topics for human review.
TAR is not meant to replace standard review processes and protocols, but instead to help streamline those processes so that review can be more targeted, fruitful and efficient.
EDRM and the Bolch Judicial Institute at Duke Law are seeking comments from the bench, bar, and public on a preliminary draft of Technology Assisted Review (TAR) Guidelines.
Technology-Assisted Review (TAR) is a concept-based method of document coding that leverages machine-learning techniques with the input of human reviewers to automate the review process.
During my [Bill Dimm] presentation at the NorCal eDiscovery & IG Retreat, I challenged the audience to create keyword searches that would work better than technology-assisted review (predictive coding) for two topics. Half of the room was tasked with finding articles about biology (science-oriented articles, excluding medical treatment) and the other half searched for articles about current law (excluding proposed laws or politics). TAR beat keyword search across the board for both tasks.
Ralph Losey’s TAR Course has a new class, the Seventeenth Class: Another “Player’s View” of the Workflow. Several other parts of the course have also been updated and edited. The TAR course now has eighteen classes.
With recent eDiscovery provider announcements that highlight the use of the terms “Continuous Active Learning” and “CAL”, provided below is a quick review and current update on Recommind’s (Recommind, Inc.) opposition to the trademark/service mark application by Maura Grossman and Gordan V. Cormack for trademarking CONTINUOUS ACTIVE LEARNING and CAL.
Special Master Maura Grossman recently issued an Order crafting a new validation protocol In Re Broiler Chicken Antitrust Litigation, (Jan. 3, 2018), which is currently pending in the Northern District of Illinois.
In each round of training, including the first, document reviewers are only looking at a few hundred, to at most, a few thousand documents. A typical complex project takes about ten rounds of machine training to complete by finding all of the relevant documents required.
In a monumental study, Fernández-Delgado and colleagues tested 179 machine learning categorizers on 121 data sets. They found that a large majority of them, were essentially identical in their accuracy.