Language-based analytics: An innovative tool for cutting e-discovery costs.
As data storage has become increasingly digital and the volume of electronically discoverable documents has grown, the problems of runaway costs, escalating strain on time and resources for e-discovery have also grown seemingly exponentially. At the same time, law departments are under increasing pressure to provide more efficient and cost-effective service to their businesses and reduce legal spend associated with litigation. Thus, e-discovery, as a major driver of litigation costs, represents a significant opportunity for streamlining and cost-cutting.
As a first line of cost-cutting defense, companies should work with their litigation counsel to build in ways to narrow the scope of e-discovery early on in cases either through discovery planning or motion practice as necessary.
Absent these tactics, companies should also consider the use of language-based analytics (LBA) in the e-discovery collection and review process. LBA can reduce the initial document review pool to a fraction of its original size by identifying and filtering out the clearly irrelevant documents. The review can then be completed relatively quickly, and with a high degree of statistical accuracy, by a relatively small group of reviewers. At bottom, LBA can dramatically increase the efficiency of e-discovery and deliver major cost savings for clients.
LBA operates based on one relatively simple, yet obvious tenet: The only truly effective way to reduce the cost and burden of e-discovery is to review fewer documents. This is accomplished through two broad workflows: language-aggregation and filtering; and smart review. An LBA system first extracts all terms used in an entire document collection, aggregates them and sorts them according to frequency of use. This provides a valuable overview of the actual language used in the documents. The overview is then reviewed with the client or others knowledgeable about the facts and issues in the case. This step, unlike systems relying on artificial intelligence, provides critical human insight in understanding the meaning of language in the context of the case.
For example, say a party has conference rooms named after former U.S. Presidents. Without identification and human review of the terms "Jefferson" and "Lincoln" in the document pool, their contextual meaning would be lost and relevant communications referring to meetings may be overlooked.
This initial language-aggregation step gives counsel a foundation on which to build their articulations of the key issues in the case, rather than making guesses or assumptions as to what words and phrases indicate relevant documents. It also allows attorneys to identify, from a concrete list of choices, frequently-used terms that are clearly relevant and, even more importantly, cannot possibly be relevant. The system then removes all documents containing only irrelevant terms from the review pool.
Although LBA is only beginning to be adopted, users report that this initial filtering can cut the review pool by 90 percent or more in complex cases. Thus, by reducing the number of documents to a fraction of the initial pool, LBA inherently reduces the drain on time and resources, and accordingly the costs, associated with document review.
RELATED STORIES: Proportionality and preservation: Getting the horse in front of the cart Experts discuss social media, Cyturf wars' and more in e-discovery webcast As wearable technology data grows, companies need clear legal hold strategy
Once filtered, the remaining documents are then subjected to the smart review phase of LBA. The smart review process improves upon predictive coding algorithms by allowing reviewers essentially to design and refine the criteria for selecting relevant documents as the review progresses. Working with the filtered pool of documents, reviewers identify specific words, phrases, figures or other data that relate to one or more of the key issues in the case. The system then generates Boolean searches based on this language, which it applies to all other documents in the pool. Documents containing similar language are automatically marked as relevant and removed from the pool, further streamlining the review process.
Critically, the LBA model compares the actual language itself rather than first relying on comparing the documents as a whole, as in predictive coding. This provides greater precision in matching and allows the system to deem documents with matching language relevant automatically rather than merely "suggesting" them to the reviewer. Moreover, because the Boolean searches are transparent and generated directly from reviewers' selection of language, they can be adjusted as necessary, unlike predictive-coding algorithms. If a reviewer's selection of a particular term or phrase causes an overly-broad range of documents to be deemed relevant, for example, that selection can be easily identified and modified to eliminate the false positives.
As a result of LBA functionality, a review team of a few as three attorneys can effectively filter and review well over a million documents in a matter of days.
Finally, the LBA process can be engineered to provide virtually any level of statistical certainty necessary to ensure that the review is defensible if challenged. An LBA system provides a test sample of documents deemed irrelevant that counsel may check periodically to verify that relevant documents are not being eliminated. This verification review can be performed as often as necessary, on as broad a sample as desired, to achieve a given confidence level.
For example, the LBA review process can be designed to achieve 95 percent certainty that all relevant documents have been produced. For incoming document reviews, of course, the same process would provide 95 percent certainty that all relevant documents have been selected and highlighted.
Although LBA is a relative newcomer to the e-discovery scene, early results have been positive to date in terms of both reducing the time and cost associated with document review and managing the scope of electronic data searching. LBA also does better on these metrics in comparison to its highly-publicized alternative: predictive coding. In sum, companies looking for ways to streamline the e-discovery process and save significant sums of money along the way (without incurring the risks currently associated with predictive coding) should consider implementing LBA in their next case.
|Printer friendly Cite/link Email Feedback|
|Publication:||Inside Counsel Breaking News|
|Date:||Dec 10, 2014|
|Previous Article:||Pro-employer ruling from Supreme Court on Integrity Case reaffirms compensation practices.|
|Next Article:||There is no CyI' in Cyteam' but there are three in Cylitigation': Selecting a winning trial team.|