Old but Still Unsolved Search Challenges: Dr. Jeremy Pickens, OpenText + Catalyst Chief Data Scientist, Wins ECIR 2019 Industry Day Best Presentation Award

By on . Posted in Seminars

Information retrieval is the science of searching for (and finding) information relevant to a user need.  The most common, most visible application of information retrieval science is found in modern web search engines, though the roots of the field extend decades before the web sprung into existence and encompass wider varieties of information needs than are typically found on the web.

The continued evolution of the field was on full display during the recent 41st European Conference on Information Retrieval (ECIR), held April 14-18 in Cologne, Germany. ECIR, which is now in its fifth decade, is a major get-together of leading information retrieval (IR) experts and data scientists from academia as well as companies including Microsoft, Google, Salesforce, IBM and others. Continue reading

Yee Haw! Texas Lawyers Required to be Competent in Technology Under Revised Rule 1.01

A big yee-haw goes out to the Texas Bar as they recently became the thirty-sixth state in the Union to codify what the ABA Model Rules of Professional Responsibility did in 2012 and that is to add very specific language about attorney competency and technology. In the newly revised Texas Rule 1.01, Paragraph 8, it states:

Because of the vital role of lawyers in the legal process, each lawyer should strive to become and remain proficient and competent in the practice of law, including the benefits and risks associated with relevant technology. Continue reading

Is Mandatory TAR on the Horizon?

Is  mandatory technology-assisted review on the horizon? This was question that Tom Gricks and John Pappas posed in a recent Bloomberg article.

Cost continues to be the primary issue lying at the heart of e-discovery disputes, particularly with the amendments to the Federal Rules of Civil Procedure (FRCP), specifically Rule 26, mandating that the scope of discovery in litigation be proportional to the value and needs of the case. Judges are increasingly being called upon to resolve these disputes and explicitly consider the fact that the cost and efficiency of different review techniques and technologies can still vary widely. Continue reading

TAR 2.0: Using Contextual Diversity to Find Out What You Don’t Know (and a Bit About Zipf’s Law)

In the early technology assisted review (TAR 1.0) era, many thought training with randomly selected documents was important to the success of a TAR review. The fear was that attorneys would bias the TAR algorithm if they selected documents for initial training. When challenged, most relied on the old shibboleth: You don’t know what you don’t know.

Frankly the “don’t know” point made sense at a time when legal professionals were just realizing that their carefully crafted keyword searches were failing because their targets used different terms. If lawyer keywords couldn’t be trusted, why trust lawyer-selected training seeds? And, random selection was the cornerstone of the TAR 1.0 protocol called simple passive learning. The computer passively (randomly) selected all the training documents. Continue reading

Legal Hold Obligations and Automation in Four Minutes

In the day-to-day world of legal departments, it can be challenging to ensure legal holds are done correctly; everything from getting the correct wording in the hold document, understanding the data types and where it exists, securing the acknowledgment, and reminding custodians of their ongoing obligation. As a result, ask a room of legal professionals, and the vast number of them will say they manage the hold process with a spreadsheet. This is not only inefficient; it leaves corporations open to a lot of unnecessary risk.

In this post, we will unravel the litigation hold problem, focus on what a litigation hold really is, where the duty comes from, and how to ensure it’s done right in a repeatable, defensible, and automated fashion. Continue reading

The New AI Executive: 6 Must-Reads for Legal

The recent Executive Order on artificial intelligence (AI), though directed at federal agencies to prioritize AI investment in research and development, is likely to continue to spur the conversation on use of AI and machine learning in the legal realm.

This is particularly so in e-discovery, where technology-assisted review (TAR), a form of AI, is seeing greater acceptance and refinement in the legal space—that is, helping corporate legal departments take control of review costs and enabling law firms to provide superior and differentiated services to their clients.

But with a deeper understanding of the technology, distinction between TAR 1.0 and TAR 2.0 systems like Catalyst’s Insight Predict (based on the continuous active learning, or CAL, protocol), and advancements to take maximum advantage of TAR techniques on more review tasks, AI can be even more useful and effective in the legal world.    Continue reading

How to Create a Knowledge-Driven Discovery Business—While Containing Costs

Corporate legal department management is quickly changing from a legal to a business process. Legal professionals, along with their business counterparts, are looking critically at how to control costs and meet ever-tightening budgets. Gone are times wistfully referred to by outside counsel as “the salad days,” where the only cost controls law departments put in place were case reserves and ever-expanding litigation budgets. Running the legal department like the rest of the corporate business units is now the rule, not the exception.

With litigation costs—and particularly discovery and document review—comprising larger and larger shares of spend, this is a ripe area to impose cost controls. However, more often than not, even the most forward-thinking in-house legal professionals don’t have the tools or insight to know where they can improve in discovery spend. Rather, most of this information resides with any number of disparate e-discovery vendors and law firms, making it near-impossible to make real-time, data-driven decisions. Continue reading

OpenText Buys Catalyst Repository Systems, Inc.

Mark J. Barrenechea, OpenText Vice Chair, Chief Executive Officer and Chief Technology Officer, talks about the acquisition of Catalyst and how the combined set of solutions will help help corporate legal departments and law firms seize the opportunities of automation, digital transformation, AI and machine learning (ML).

“AI, analytics and ML are disrupting traditional approaches, empowering corporations to take better control of their eDiscovery processes and costs and law firms to provide superior, differentiated services for their customers,” says Barrenechea. “With legal solutions available as hosted (SaaS) or managed services, OpenText is uniquely positioned to help corporations take advantage of market disruption. We will continue to invest in eDiscovery processes, analytics, AI, cloud and managed services to improve effectiveness, deliver insights and maximize existing technology investments for our customers.”

Read the OpentText blog post.

And the 2018 Award Goes to… TAR 2.0

By on . Posted in TAR 2.0

As an annual tradition, we compile a list of the most widely read Catalyst blog posts of the previous year to see what topics most interest our readers. Here are our top five most popular blog posts of 2018.

1. 57 Ways to Leave Your (Linear) Lover

What’s more fun than running 57 simulations for a client investigation? Seeing the results.

We structured a simulated review on Insight Predict, our TAR 2.0 platform, to be as realistic as possible, looking at the client’s investigation from every conceivable angle. The results were outstanding, so we ran it again, using a different starting seed. We did 57 different simulations starting with relevant seeds (singularly with each relevant document), a non-relevant seed and a synthetic seed. Regardless of the starting point, Predict was able to locate 100% of the relevant documents after reviewing only a fraction of the collection.

Continue reading

Redactions Done Right: How to Avoid the Manafort Fiasco

The attorneys for former Trump campaign chairman Paul Manafort appear to have made a rather significant mistake in a court filing earlier this week, allowing redacted portions to be revealed by copying and pasting the text to a new document. The filing reveals alleged campaign communications between Manafort and Russian operative in which he shared polling data during the campaignan allegation that Manafort earlier denied to federal investigators.

While an essential component of litigation review, redaction mistakes are easy to make, as it’s cumbersome and time-consuming to go through a document and draw black boxes over individual words and phrases. When redaction missteps occur, it can mean trouble for attorneys and their clients—including running afoul of various applicable rules including the American Bar Association’s rule on confidentiality. Continue reading