Legal Technology News - E-Discovery and Compliance Blog

Search Feed

March 12, 2013

EDI-Oracle Study Update

It has been a few weeks since our last update. We have extended the time to respond for the participants. While some thought the extension was unfair to those that have submitted on or before the due date, Pallab, the professors, and I all want as many data points as possible in the evaluation. That said, we are taking note of those teams that submitted early.

Again, we should remind all participants that time is one of the factors in the subsequent evaluation by Oracle. Therefore, a quicker turnaround is preferable and no participant should withhold submitting results unless the additional time is required to complete the review. While there is no penalty for submitting after the first deadline, Oracle will look favorably upon those that finish quickly.

Also, if there were tasks required by the protocol that your firm would not have normally undertaken in a real-life review (categorizing privilege for NR docs), please identify that task as a break-out on your model invoices with an explanation. You may submit your model invoices at you leisure, but no later than May 1.



March 04, 2013

Search Software: 25 Things to Consider

4-levels-pyramid Selection of advanced TAR or CAR-based type search and review software is not an easy decision.

I had a chance recently to study the private manuscript of Professor Doug Oard on search design, Information Retrieval for E-Discovery. He is one of, if not the top independent expert on search in the world today. His manuscript written with William Webber is a survey that will be used to teach the next generation of information retrieval engineers, the ones who will help design the legal search software of the future.

Doug was also kind enough to try to explain key portions of this manuscript to me. The end result is a rather complex and lengthy blog entitled The Many Types of Legal Search Software in the CAR Market Today. Here is my takeaway list of twenty-five plus factors you should consider to evaluate the strength of search software. They should help you to evaluate the overall quality, effectiveness, and efficiency of the software on the market today.

  1. Search of content.
  2. Search of contextual metadata.
  3. Search of descriptive metadata.
  4. Search of behavioral metadata.
  5. Classifier design.
  6. Active query formulation features (building blocks).
  7. Annotated example training features (pearl growing).
  8. Random training abilities.
  9. Human judgment training features.
  10. Software uncertainty selected training.
  11. Balance in your classifier system between complexity and intuitiveness.
  12. Deduplication, near deduplication, and similarity searches.
  13. Concept searches, what exactly are they doing and how.
  14. Clustering features.
  15. Email thread and families.
  16. Quality control features.
  17. Quality assurance features.
  18. Confidentiality protection features, including privilege logging and redacting.
  19. Recommended method(s) of use (level four).
  20. Ease of operation.
  21. Clarity and efficiency of user interface.
  22. Customizability for particular licensees, projects, users.
  23. Response time (each entry and length of search processing).
  24. Reporting capacities and usefulness.
  25. Expense.

There are many other criteria, I know, such as information security, more specific user interface issues, reviewer and production issues, reviewer monitoring and interface, online access, etc. Software selection should also carefully consider the software support and training issues. Project management support is also critical, some say just as important as the software itself, especially for novice users of the search software.

November 07, 2012

USPTO Grants H5 Patent for Its TAR Process

H5_logo_400We interrupt your construction of a standard query designed to retrieve relevant documents, with high recall and precision, from a corpus of documents based on your expertise as an attorney and content expert in a case and your iterative work at refining a search query with reflexive search terms after your review of search results based on previous, iterative searching. Attention.

H5, a provider of e-discovery, technology-assisted review and case preparation support, has patented a process for "high recall and high precision relevancy searching." The U.S. Patent and Trademark Office granted to H5 on October 23 U.S. Patent No. 8,296,309, which covers aspects of H5's "deterministic" technology-assisted review, according to the company — and the USPTO. We checked.

Continue reading "USPTO Grants H5 Patent for Its TAR Process" »

August 31, 2012

Resistance Is Not Futile

IMG_3150I'm home with a broken ankle today and so was able to watch the stock market reaction to the release of Federal Reserve Chairman Ben Bernanke's speech on fiscal policy. Tens of thousands of brokers and investors were eagerly awaiting this speech to try to forecast the direction of the stock market. Many were ready to immediately execute trades based on the content of this writing.

I watched CNBC and the market reaction to the speech. The market almost immediately dropped 100 points, just minutes after the speech was released to the public and before he had even begun reading his speech. No human could possibly have read his speech that fast. At best they could have skimmed a few paragraphs. Yet millions, perhaps billions of dollars changed hands as the market plummeted. Then, as if by magic, in the next hour the market climbs back up that 100 points lost, and goes even higher. What the hell happened? Bernanke did not depart from his published speech.

A CNBC reporter on the floor of the New York Stock Exchange (I think it was Bob Pasini) provided a very interesting explanation, one which I predict will not be again repeated and may even be hushed up. I will not forget it because it made perfect sense to me and has implications for legal search in e-discovery.

Read the full commentary on LTN online.

Image: courtesy of Ralph Losey

April 17, 2012

Speaking of 'Da Silva Moore' — Recusal?

Peck_andrew2_400"The e-discovery dispute in Monique da Silva Moore, et al. v. Publicis Group SA, et al. (Case No. 11-CV-1279), in the U.S. District Court for the Southern District of New York, took one step closer to a reality show on April 13, when plaintiffs filed a formal motion to recuse or disqualify Magistrate Judge Andrew Peck," writes Sean Doherty.

Read the full article, "E-Discovery Dispute Yields Formal Recusal Request in 'Da Silva Moore,'" on LTN online.

Image by Monica Bay

Predictive Coding Watch: ‘Kleen Products’ in Illinois

Speed_checkMeanwhile, back at the ranch ... With all eyes on United States District Court  Magistrate Judge Andrew Peck's orders involving "computer-assisted review" and plaintiffs' challenges in Da Silva Moore in the Southern District of New York, another decision weighing the automated review technology, Kleen Products, LLC, et. al. v. Packaging Corporation of America, et. al., is quietly playing itself out in the Northern District of Illinois.

As related in a post by Matthew Nelson, senior e-discovery counsel at Symantec, on the e-discovery 2.0 blog, this time out it's the plaintiffs challenging the defendants' discovery methods because they didn't use predictive coding. In a federal antitrust case before Magistrate Judge Nan Nolan, the plaintiffs attorneys are requesting that the defendants redo production. As Nelson describes it the plaintiffs' argument appears to be "that defendants should have used predictive coding to avoid the limitations of keyword search tools." 

Symantec owns Clearwell Systems, and its e-discovery software is being used by at least one of the defendants, according to the transcripts of the first evidentiary hearing in February. With review 99% complete and more than a million documents already produced, Nelson (remember, he's not exactly an disinterested party here) questions the timing of the plaintiffs challenge.

Is this merely a bid by plaintiffs to have the other side pay for the plaintiffs' culling and filtering? With a second hearing completed March 28 and a third day of hearings scheduled, Da Silva Moore isn't the only show running.

For more on the two cases, check out the eDiscovery Journal and the Document Review MD Blog.

Image by

October 04, 2011

Judge Peck: Green Light for Computer-Aided Coding

PeckIn the October issue of Law Technology News magazine, New York-based U.S. Magistrate Judge Andrew Peck argues that the time has come for computer-assisted coding — a.k.a. predictive coding — to be embraced by litigators and the judiciary. 

In "Search, Forward," Peck chronicles the evolution of search, concluding lawyers' fears of judicial rebuffs or potential Daubert hearings "seem largely misplaced." When using computer-assisted coding, he suggests that litigators should be prepared to explain "what was done and why that produced defensible results."  

"Until there is a judicial opinion approving (or even critiquing) the use of predictive coding, counsel will just have to rely on this article as a sign of judicial approval. In my opinion, computer-assisted coding should be used in those cases where it will help 'secure the just, speedy, and inexpensive' (Fed. R. Civ. P. 1) determination of cases in our e-discovery world."

Image: Russ Curtis

Sign Up for the E-Discovery and Compliance Newsletter

An Affiliate of the Network

From the Newswire

Sign up to receive Legal Blog Watch by email
View a Sample

Contact EDD Update

Subscribe to this blog's feed

RSS Feed: LTN Podcast

Monica Bay's Law Technology Now Podcasts are also available as an RSS feed.

Go to RSS Subscribe page

March 2013

Sun Mon Tue Wed Thu Fri Sat
          1 2
3 4 5 6 7 8 9
10 11 12 13 14 15 16
17 18 19 20 21 22 23
24 25 26 27 28 29 30

Blog Directory - Blogged