TAR Unplugged: Live Q&A with the Experts

Transcription

1 catalystsecure.com E-Discovery Webinar TAR Unplugged: Live Q&A with the Experts Andrew Bye Thomas Gricks

2 Presenters Thomas C. Gricks III Andrew Bye Managing Director Director, Machine Learning Catalyst Catalyst

3 Managing Today s Webinar We will start with questions from previous webinars We will handle three main topics sequentially Terminology Process Legal Listeners can ask topical questions at any time using the Q&A function

4 Terminology Can you What is the explain practical the difference difference between a TAR seed 1.0set, anda TAR training 2.0? set and a control set?

5 TAR 1.0

6 Control Set

7 Seed Set

8 Training Set

9 TAR 2.0

10 Training Set

11 Terminology Whatseem You doesto TAR be using reviewed Responsive mean? and Relevant Does that imply interchangeably. that a reviewer Is has there a at looked difference? the document?

12 Process

13 Process How many How What Can TAR can if some handle documents avoid ofthe my bias huge custodians will infiles the I have beginning liketohave Excel? review do I Iknow TAR process is in been How of What Can athe TAR TAR you would keyword steps review? project? comment bethat do used If culled I At work? need I m what for on not and multi-language to the What point starting take others impact about can to pull with you of the a working the documents are I d like a and better clarification as tobeing how provide haven t non-text large plug data deep on sets? random neural a afiletypes? reliable will TAR Would networks that set review estimate? of affect you documents, and for have the document Is be TAR to there assured make won t a way a correctly ranked during a review? richness is initially determined? to know process? the that separate classification? initial my that results model training What before are about for be you defensible? each skewed? date start? language? culling?

14 Legal What are some major TAR rulings that I should know about when planning a review?

15 Using TAR 1. (2012) da Silva Moore v. Publicis Groupe Agreed on predictive coding; disagreed on protocol 2. (2012) Global Aerospace v. Landow Aviation Judge permitted predictive coding over objection 3. (2015) Rio Tinto Plc v. Vale S.A. ( Black Letter Law ) 4. (2016) Hyles v. New York City Cannot compel the use of TAR** 5. International Irish Bank Resolution Corp. v. Quinn, (Ireland, 2015) Brown v. BCA Trading; Pyrrho Invmts. v. MWB Property, (England, 2016) Money Max v. QBE Insurance Group, (Australia, 2016)

16 Advance Keyword Culling 1. (2013) In re Biomet M2A Magnum Hip Implant Products Liability Litigation Keyword culling was permitted Context: 19.5MM 3.9MM 2MM TAR 2. (2017) FCA US v. Cummins Keyword culling was denied Context: Advance ESI protocol negotiation

17 Recall Thresholds 1. (2012) Global Aerospace v. Landow Aviation Accepted protocol for 75% recall 2. (2014) Independent Living Center v. City of Los Angeles The Court is anticipating that the quality assurance will establish that the recall is 75%. If the percentage is lower than 75%, counsel should bring that issue to the Court s attention.

18 Q&A Thomas C. Gricks III Managing Director, Professional Services Catalyst Tom is a managing director at Catalyst and a licensed attorney in Pennsylvania. Before joining Catalyst, he was a general commercial litigator for 23 years. He practiced before both federal and state courts across the United States, and, in addition to Pennsylvania, is admitted to practice in a number of federal jurisdictions, including the Supreme Court. For the past several years, Tom has devoted a substantial portion of his practice to e-discovery, with a particular emphasis on technology assisted review (TAR). He argued the Global Aerospace case before the Circuit Court in Loudoun County, Virginia, which is the first case to permit the use of TAR over the objection of the opposing party. Andrew Bye Director, Machine Learning & Analytics Catalyst Andrew is the director of machine learning and analytics at Catalyst, and a search and information retrieval expert. Throughout his career, Andrew has developed search practices for e-discovery, and has worked closely with clients to implement effective workflows from data delivery through statistical validation. Before joining Catalyst, Andrew was a data scientist at Recommind. He has also worked as an independent data consultant, advising legal professionals on workflow and search needs. Andrew has a bachelor s degree in linguistics from the University of California, Berkeley and a master s in linguistics from the University of California, Los Angeles.

19 Thank You Find more resources at: catalystsecure.com/resources