Faced with the accelerating increase in the volume of Electronically Stored Information (ESI) and the emergence of the concept of Big Data, enterprises worldwide need next generation IT systems to fulfill their corporate compliance, information governance and eDiscovery requirements to process and analyze all of this data. It is in response to this demand that Technology Assisted Review (TAR) (also known as Predictive Coding or Computer Assisted eDiscovery) is emerging as a viable and recognized IT option.
Enterprises worldwide face ever-increasing challenges to provide next generation IT systems to fulfill the corporate compliance, information governance and eDiscovery requirements to process and analyze this data. eDiscovery itself consists of the identification, collection, analysis and production of ESI that is potentially relevant evidence for a legal matter. Industry experts generally agree that approximately 70% of the overall cost of eDiscovery is the cost of lawyers and paralegals manually reviewing documents to determine whether or not they are relevant to the legal matter in question.
As a result, enterprises utilize next generation eDiscovery technology such as Early Case Assessment (ECA) software to initially cull identified and collected large datasets. Using ECA software, enterprise can reduce the data down to a more manageable and potentially relevant dataset that ultimately reduces the cost of the manual document review stage. However, enterprises still seek next generation technology to reduce eDiscovery costs and other related Big Data analysis requirements such as Information Governance even further.
Computer Assisted eDiscovery (CAeD) utilizes advanced search, text analytics, and statistical modeling to perform document identification tasks at the same level of competence as human review. Already great strides have been made in this area as a recent article entitled, “Technology-Assisted Review in eDiscovery Can Be More Effective and More Efficient Than Exhaustive Manual Review” co-written by Dr. Maura Grosmann and Dr. Gordon Cormack conclude that TAR could in fact be more effective than human review.
As an example, eDiscovery software such as Axcelerate, from Recommind, utilizes advanced predictive analytics to automatically prioritize documents into categories based on concepts which provides litigators more usable information than was previously allowed with keyword search.
In another example, Zoom, from Equivio enables experienced litigators to train the software to indentify relevant documents in a statically significant manner in the same way that they would manually. In both cases, users report that Axcelerate and Zoom reduced the amount of time and resulting cost of human document review.
As important as CAeD’s ability to programmatically perform these tasks, legal precedence has already been set recognizing CAeD as a viable and recognized alternative to human review. On February 24, 2012 in Da Silva Moore v. Publicis Groupe & MLs Group, No. 11 Civ. 1279 (ALC)(AJP) (S.D.N.Y. Feb. 24, 2012), Magistrate Judge Andrew Peck issued an opinion approving the use of TAR for ESI, making it the first Federal case ever to recognize TAR as an acceptable way to identify relevant ESI.
This was not without controversy. While Peck was not faced with the issue of two parties arguing about the propriety of TAR in general (the parties had agreed to the use of TAR at the outset) they did struggle with defining a mutually-agreeable protocol.
Plaintiffs objected to the ruling claiming that the court-approved TAR protocol “risks failing to capture a staggering 65% of the relevant documents in this case.” Plaintiffs even went so far as to ask Judge Peck to recuse himself because of an alleged relationship with defense counsel and the vendor selected for the case. (This request was denied and later affirmed.)
In another case in March, 2012, in Kleen Products LLC v. Packaging Corporation of America, et al., Magistrate Judge Nan Nolan sitting in the United States District Court for the Northern District of Illinois faced plaintiffs who asked that defendants be ordered to redo part of their production and use TAR in lieu of keyword searches.
Relying on the Sedona Principles, Judge Nolan confirmed that opposing parties cannot dictate what technology solutions their opponents must use, and that “the responding parties are better suited to evaluate the procedures, methodologies, and technologies appropriate for preserving and producing their own [ESI].” In August 2012, only after months of debate, did the parties reached an agreement to drop the TAR-based approach, at least for the first request for production.
However TAR continues its almost inevitable march to broader legal acceptance and adoption. On April 23, 2012, a Virginia state court case, Global Aerospace Inc., et al, v. Landow Aviation, L.P. dba Dulles Jet Center, et al, Consolidated Case No. CL 61040, Circuit Court of Loudoun County, became the first in which TAR was actually ordereddespite the plaintiff’s objections that the technology was not as effective as purely human review. Karl Schieneman Esq., President of eDiscovery company Review Less is on the team that helped Schnader Harrison Segal & Lewis LLP choose OrcaTec’s OrcaPredict as the predictive coding to use in the Global Aerospace case.
“The lack of seed sets, the ease of use, the functionality and the sheer brainpower behind OrcaPredict, including having Herb Roitblat as a potential expert witness, made it the right choice for Schnader in such an important case,” said Schieneman. “We have all been very pleased with the way it is working.”
The evidence is becoming indisputable. Despite plaintiff objections, courts are concluding that CAeD is a viable option for legally defensible eDiscovery. More notably, they are at this point not recommending or legislating any specific type of CAeD software such as Axcelerate or Zoom to be used in these eDiscovery efforts. Instead the courts are leaving that decision to the discretion of the parties involved. What they are doing is recognizing the validity of using CAeD as a viable means of performing eDiscovery so that enterprises may reduce the amount of time and resulting cost of human document review.
- How to Engage Enterprise Buyers in Meaningful Conversations in 2016 February 28, 2016
- nVIDIA Driving Deep Learning to the Forefront – Literally February 22, 2016
- New Technologies Disrupting the Legal Business in the UK February 17, 2016
- Shares of Tableau plunge 36% after company posts $41M loss in Q4 February 5, 2016
- LexisNexis Unveils Lexis® DiscoveryIQ eDiscovery Platform Enhanced by Brainspace February 2, 2016