The TAR Course has a new class, the Seventeenth Class: Another “Player’s View” of the Workflow. Several other parts of the Course have been updated and edited. It now has Eighteen Classes (listed at end). The TAR Course is free and follows the Open Source tradition. We freely disclose the method for electronic document review that uses the latest technology tools for search and quality controls. These technologies and methods empower attorneys to find the evidence needed for all text-based investigations. The TAR Course shares the state of the art for using AI to enhance electronic document review.
The key is to know how to use the document review search tools that are now available to find the targeted information. We have been working on various methods of use since our case before Judge Andrew Peck in Da Silva Moore in 2012. After we helped get the first judicial approval of predictive coding in Da Silva, we began a series of several hundred document reviews, both in legal practice and scientific experiments. We have now refined our method many times to attain optimal efficiency and effectiveness. We call our latest method Hybrid Multimodal IST Predictive Coding 4.0.
The Hybrid Multimodal method taught by the TARcourse.com combines law and technology. Successful completion of the TAR course requires knowledge of both fields. In the technology field active machine learning is the most important technology to understand, especially the intricacies of training selection, such as Intelligently Spaced Training (“IST”). In the legal field the proportionality doctrine is key to the pragmatic application of the method taught at TAR Course. We give-away the information on the methods, we open-source it through this publication.
All we can transmit by online teaching is information, and a small bit of knowledge. Knowing the Information in the TAR Course is a necessary prerequisite for real knowledge of Hybrid Multimodal IST Predictive Coding 4.0. Knowledge, as opposed to Information, is taught the same way as advanced trial practice, by second chairing a number of trials. This kind of instruction is the one with real value, the one that completes a doc review project at the same time it completes training. We charge for document review and throw in the training. Information on the latest methods of document review is inherently free, but Knowledge of how to use these methods is a pay to learn process.
The Open Sourced Predictive Coding 4.0 method is applied for particular applications and search projects. There are always some customization and modifications to the default standards to meet the project requirements. All variations are documented and can be fully explained and justified. This is a process where the clients learn by doing and following along with Losey’s work.
What he has learned through a lifetime of teaching and studying Law and Technology is that real Knowledge can never be gained by reading or listening to presentations. Knowledge can only be gained by working with other people in real-time (or near-time), in this case, to carry out multiple electronic document reviews. The transmission of knowledge comes from the Q&A ESI Communications process. It comes from doing. When we lead a project, we help students to go from mere Information about the methods to real Knowledge of how it works. For instance, we do not just make the Stop decision, we also explain the decision. We share our work-product.
Knowledge comes from observing the application of the legal search methods in a variety of different review projects. Eventually some Wisdom may arise, especially as you recover from errors. For background on this triad, see Examining the 12 Predictions Made in 2015 in “Information → Knowledge → Wisdom” (2017). Once Wisdom arises some of the sayings in the TAR Course may start to make sense, such as our favorite “Relevant Is Irrelevant.” Until this koan is understood, the legal doctrine of Proportionality can be an overly complex weave.
The TAR Course is now composed of eighteen classes:
- First Class: Background and History of Predictive Coding
- Second Class: Introduction to the Course
- Third Class: TREC Total Recall Track, 2015 and 2016
- Fourth Class: Introduction to the Nine Insights from TREC Research Concerning the Use of Predictive Coding in Legal Document Review
- Fifth Class: 1st of the Nine Insights – Active Machine Learning
- Sixth Class: 2nd Insight – Balanced Hybrid and Intelligently Spaced Training (IST)
- Seventh Class: 3rd and 4th Insights – Concept and Similarity Searches
- Eighth Class: 5th and 6th Insights – Keyword and Linear Review
- Ninth Class: 7th, 8th and 9th Insights – SME, Method, Software; the Three Pillars of Quality Control
- Tenth Class: Introduction to the Eight-Step Work Flow
- Eleventh Class: Step One – ESI Communications
- Twelfth Class: Step Two – Multimodal ECA
- Thirteenth Class: Step Three – Random Prevalence
- Fourteenth Class: Steps Four, Five and Six – Iterative Machine Training
- Fifteenth Class: Step Seven – ZEN Quality Assurance Tests (Zero Error Numerics)
- Sixteenth Class: Step Eight – Phased Production
- Seventeenth Class: Another “Player’s View” of the Workflow (class added 2018)
- Eighteenth Class: Conclusion
With a lot of hard work you can complete this online training program in a long weekend, but most people take a few weeks. After that, this course can serve as a solid reference to consult during complex document review projects. It can also serve as a launchpad for real Knowledge and eventually some Wisdom into electronic document review. TARcourse.com is designed to provide you with the Information needed to start this path to AI enhanced evidence detection and production.
Reblogged this on Legal Career Center.