Technology Assisted Review For UK Lawyers

TECHNOLOGY - ASSISTEDREVIEW · AGUIDEFORUKLAWYERS Technology-Assisted Review for UK Lawyers Predictive Coding, Continuous Active Learning and Defending the Machine's Judgement COMPUTERFORENSICSLAB § ABOUTTHEAUTHOR PRE PA REDBYCOMPUTERFORENSICSLABE - DISCOVERYTEAM VALIDATION STATISTICS CPR PART 35 EXPERT REPORTS FULL CHAIN-OF-CUSTODY DOCUMENTATION § CONTENTS In this guide 01 Executive summary 02 The problem in plain English: teaching a machine what relevant looks like 03 How TAR works: models, training and the two generations 04 The human layer: seeds, subject-matter reviewers and consistency 05 Validation: recall, precision and the stopping decision 06 TAR in the protocol: disclosure, negotiation and the courts 07 TAR in the toolchain: keywords, threading and privilege 08 Source architecture: where else the evidence lives 09 Worked examples 10 Common mistakes and technical limitations 11 Questions to ask · Suggested wording 12 Checklist and red flags · When to involve a digital forensic expert 13 Frequently asked questions 14 Glossary · References · Disclaimer · How a specialist laboratory can assist § 0 1 · ORIENTATION Executive summary THEHEADLINEPOINT : TARISHUMANJUDGEMENTAMPLIFIEDBYRANKING : ITS DEFENSIBILITYISTHETRAININGRECORDPLUSTHEVALIDATIONSTATISTICS , ANDBOTH AREBUILT , NOTASSUMED § 0 2 · FIRSTPRINCIPLES The problem in plain English: teaching a machine what relevant looks like § 0 3 · THEMACHINERY How TAR works: models, training and the two generations § 0 4 · THETEACHERS The human layer: seeds, subject-matter reviewers and consistency § 0 5 · THELICENCETOSTOP Validation: recall, precision and the stopping decision § 0 6 · THEPAPERWORKANDTHECOURTS TAR in the protocol: disclosure, negotiation and the courts § 0 7 · THETOOLCHAINSEAT TAR in the toolchain: keywords, threading and privilege § 0 8 · THEWIDERMAP Source architecture: where else the evidence lives QUESTION QUALIT Y TRAINING VALIDATION PROTOCOL & PLATFORM NON-TEXT RECORD FILE CORRESPONDENCE LOGS ROUTES TEXT- LAYER § 0 9 · INTHEWILD Worked examples EXAMPLE1 · THE1 . 8MILLIONDOCUMENTSREVIEWEDBY140 , EXAMPLE2 · THESTOPPINGDECISIONTHATWASABUDGETINCOSTUME EXAMPLE3 · THEELUSIONFINDTHATWASASCOPEHOLE § 1 0 · WHEREITGOESWRONG Common mistakes and technical limitations Common mistakes Technical limitations § 1 1 · INTERROGATORIES & DRAFTINGAIDS Questions to ask · Suggested wording Ask your client Ask your opponent Ask your eDiscovery / forensic provider SUGGESTEDWORDING · TA R PA R AG RAPHFORTHEDRD / PROTOCOL § 1 2 · QUICKCONTROL Checklist and red flags · When to involve a digital forensic expert The TAR checklist Red flags When to involve a digital forensic expert § 1 3 · COMMONQUESTIONS Frequently asked questions Is predictive coding actually approved by the English courts? How big does a matter need to be before TAR makes sense? Can the model miss the smoking gun? Do we have to tell the other side we are using TAR? What recall target should we agree? Is TAR the same as using generative AI for review? § 1 4 · REFERENCE Glossary CAL Sources and authoritative references DISCLAIMER § HOWASPECIALISTLABORATORYCANASSIST Working with Computer Forensics Lab Speak to a forensic examiner, not a salesperson. INSTRUCTTHELAB NEWENQUIRIESEMAILE - DISCOVERY