Defensibility of predictive coding software

Over the past three years, the predictive coding market has transitioned from a preliminary. While predictive coding has been the dominant ai technology in ediscovery since 2012, new ai technologies are emerging, and ediscovery could be on the cusp of another. Monday, june 18ths wall street journal article on predictive coding marked the latest top tier news entry into the debate, highlighting the impact of predictive coding. Given the requirements for creating a seed set to properly execute the predictive coding. Publicis groupe and msl group opinion, magistrate judge andrew peck of the southern district of new york explicitly approved the use of predictive coding. Visual predictive coding ringtail ediscovery software. This comparison results in prediction errors that are then used to. However, it is commonly performed in a manner that may be severely underinclusiveand therefore raise concerns about its defensibility.

If you tagged a training document as relevant and asked the software to predict whether an exact duplicate or a neardupe of the very same document was relevant, how. No matter how intelligent, a computer program is ultimately a logical. Predictive coding is an effective ediscovery tool for ranking large sets of. An article called the red herring of defensibility and predictive coding by craig carpenter of recommind on the inforiskawareness site draws attention again to the technology generally known as predictive coding. Predictive coding software is a form of machine learning that takes data input by people about document relevance and then applies it to much larger document sets. This book is an attempt to catch lightning in a bottle. Initially, the global plaintiff argued that all relevant evidence must be produced, which is, after all, the goal of defensible discovery. The uncertain promise of predictive coding iowa law. There are significant benefits and risks to predictive coding, and only a few courts have had the opportunity to evaluate and balance this equation. Once biomet had reduced the document count, it employed a predictive coding tool on the culled and deduplicated data set, identifying almost. May 29, 20 quite simply, the quality of the predictive coding results is dependent upon the quality of the predictive coding input, as the software will propagate bad coding decisions as well as good coding.

There are a handful of judges whose pronouncements on anything ediscoveryrelated are bound to get legal technologists talking. The defensibility of predictive coding in ediscovery. Discoverready, predictive coding and freedom of choice. The right combination of people and transparent processes will ensure defensibility. The major defensibility issues around predictive coding can be divided into two general categories. Aug 17, 2012 top ten best practices in predictive coding predictive coding is new to ediscovery. Dont jump the gun with predictive coding warroom document.

Top ten best practices in predictive coding predictive coding is new to ediscovery. If you tagged a training document as relevant and asked the software. Perspectives on predictive coding is the best reference book available on this. Monday, june 18ths wall street journal article on predictive coding marked the latest top tier news entry into the debate, highlighting the impact of predictive coding on document intensive litigation. Publicis groupe and msl group opinion, magistrate judge andrew peck of the southern district of new york explicitly approved the use of predictive coding in electronic discovery. Misleading metrics and irrelevant research accuracy and. Advice from counsel can predictive coding deliver on its. What is predictive coding, and how does it apply to. With defensibility largely settled and comparisons showing how much more efficient a tarenabled workflow can be, its time for organizations to table any lingering objections about the process. Jan 09, 2012 predictive coding is an effective ediscovery tool for ranking large sets of documents. Our technologists bring a myriad of industry experience to collaborate with our discovery counsel to build costaligned, defensible. Workflow technology is integrated with predictive coding to automate the routing.

The right combination of people and transparent processes will ensure defensibility and visibility into the results. Predictive coding is a process that requires legal, technical and statistical expertise. Symantec white paper predictive coding defensibility. Jul 21, 2016 courts are unlikely to be persuaded as to the reliability or defensibility of a predictive coding process simply by being given a detailed explanation of how the specific algorithm operates. Built in close collaboration with ediscovery leaders in both corporations and law firms, ringtails predictive coding delivers a.

Curious about predictive coding but not sure how to set up a model. Ediscovery teams are also discovering that predictive coding can now be applied across. Judge andrew peck, united states magistrate judge for the southern district of new york is one of them. Reviewing information is the most expensive part of the ediscovery process. Legal professionals and the two challenges in predictive coding. However, it is commonly performed in a manner that may be severely underinclusive and, therefore, raises concerns about its defensibility. On february 24, in the first decision of its kind, monique da silva moore, et al. Kroll ontrack voted best predictive coding solution in new. The client can avoid the perdocument cost for the predictive coding software. Predictive coding and defensibility edisclosure information. The value of predictive coding in internal investigations the task of collecting, identifying and properly using key documents is the foundation of a successful internal investigation. Symantecs clearwell ediscovery platform to offer transparent. Dec 14, 2010 predictive coding and defensibility posted on december 14, 2010 by chris dale an article called the red herring of defensibility and predictive coding by craig carpenter of recommind on the inforiskawareness site draws attention again to the technology generally known as predictive coding. The central issue is therefore to ensure that the underlying process is well considered and the output can be verified in terms of statistical accuracy.

However, with predictive coding, the respondents overwhelmingly agreed that humans become an even more important part of the process since they determine the reference set, test and refine the software, and conduct additional quality control to ensure defensibility. This months column will address the defensibility of predictive coding in court cases, whether the use of predictive coding should be disclosed to the opposing party, and how to address challenges to the use of predictive coding. Oct 15, 2019 discoverready, predictive coding and freedom of choice. So, how should your organization begin when preparing a collection for predictive coding discovery. His recent article, search, forward, published in law technology news, is one of the few judicial pronouncements on the use of predictive coding. Predictive coding is the common name for this process which is emerging as an accepted. Remarkably, although there is significant industry concern regarding the defensibility of predictive coding and relatively little concern about the defensibility of concept analytics, it is evident that predictive coding is gaining traction at a far greater rate. David kessler of fulbright and jaworski weighed in on judge pecks recent predictive coding opinion this week, with. Precision, p, is the percentage of documents predicted to be relevant that actually are relevant. Repeatable, defensible workflow means everyday use of predictive coding. It requires an iterative approach that includes statistical sampling and quality. Many lawyers and technologists like predictive coding and recommend it to their colleagues.

Recommind contra mundum posted on june 19, 2011 by chris dale it is a novel experience to spend a whole saturday writing a 4,330word article whose conclusion is that none of its subjectmatter is really very important to ones readers, however much it means to the participants in the story. Predictive coding teams often find that, during the process of training the software. The model is used to generate predictions of sensory input that are compared to actual sensory input. Predictive coding also known as predictive processing is a theory of brain function in which the brain is constantly generating and updating a mental model of the environment. Predictive coding and defensibility posted on december 14, 2010 by chris dale an article called the red herring of defensibility and predictive coding by craig carpenter of recommind on the inforiskawareness site draws attention again to the technology generally known as predictive coding. Do not take predictive coding if you are allergic to it, i. Legal professionals and the two challenges in predictive. On the process side, a party using predictive coding will need to be prepared. The symantec transparent predictive coding workflow who should read this paper predictive coding is one of the most promising technologies to reduce the high cost of. Learn more about thoughts on whether predictive coding is defensible. Quality control and sampling monitors the effectiveness of document categorizations to arrive at the necessary confidence levels to ensure. This particular principle has been used by producing parties, along with federal rule of civil procedure. How predictive coding is changing the legal landscape.

Courts are unlikely to be persuaded as to the reliability or defensibility of a predictive coding process simply by being given a detailed explanation of how the specific algorithm operates. Ringtails integrated predictive coding gives your team everything they need to effectively analyze and reliably autocode documents. Predictive coding is an effective ediscovery tool for ranking large sets of documents. This particular principle has been used by producing parties, along with federal rule of civil procedure fre 26b1 to defend its chosen method for identifying and producing documents. Clients interactions with these technologies are mediated by lawyers who train the. Kldiscovery predictive coding is an awardwinning, patented technology that learns from both user decisions and random sampling to immediately and continuously prioritize important documents for. Leverage predictive coding and ai for defensible and effective. Apr 28, 2019 to bring technology inhouse, you have to understand the legal landscape. Grimm and judge lorna schofieldpredictive coding technology is becoming necessary in some cases and a point of conversation in several others since judge andrew peck allowed the use of predictive coding in the landmark 2012 da silva moore case. Misleading metrics and irrelevant research accuracy and f1. Predictive coding type software analyzes whole documents in a dataset, not just keywords, and uses advanced mathematics, including nearinfinitedimensional vector space probability analysis, and logistic regression algorithms, to order, compare, and rank them. In practice, despite the buzz and several generations of technology development predictive coding. What does counsel really think about predictive coding. Predictive coding software is a form of machine learning that takes data input by people.

Predictive coding is the hot ediscovery topic of 2012, with three significant cases da silva moore v. Now that the uk courts have accepted the use of predictive coding software, it is crucial that lawyers understand how to apply it and determine which type of case will benefit most. Reviewers code each document as relevant or not relevant to the case and input this information into the predictive coding software. For the last decade, predictive coding has dominated ediscovery conferences and preoccupied industry thought leadership. Predictive coding is often employed only to predict whether documents. Built in close collaboration with ediscovery leaders in both corporations and law. Predictive coding uses classifiers to extrapolate human coding decisions made on a subset of materials to a larger data set. Even if one party demonstrates that its chosen method is better. Misleading metrics and irrelevant research accuracy and f1 leave a reply if one algorithm achieved 98.

Predictive coding demonstrable results equal acceptance. We join together veteran ediscovery partners, associates, project managers, technology professionals and data analysts to offer you an integrated, streamlined approach to ediscovery. Quite simply, the quality of the predictive coding results is dependent upon the quality of the predictive coding input, as the software will propagate bad coding decisions as well as good coding. Applied this way, predictive coding functions as a culling tool. People are clamoring for a judge to embrace predictive coding. Limitations of available technology to provide a repeatable and well. Advice from counsel can predictive coding deliver on its promise. Mar 23, 2011 predictive coding is the electronic coding, organization, and prioritization of entire sets of electronically stored information esi according to their relation to discovery responsiveness. In contrast to legalforms software, predictivecoding technologies are used as lawyers tools not lawyer substitutes.

The power of kroll ontracks predictive coding platform is derived from three components that work together. There are recent court decisions endorsing the use of predictive. A curated ediscovery case law database of predictive coding, tar, cal and machine learning related. Jan 23, 2012 benefits of transparent predictive coding. Limitations on the defensibility of manual and keyword searching. A loose explanation of predictive coding is that a senior lawyer or subjectmatter expert makes relevance decisions about a subset of documents. Predictive coding electronic discovery best practices. Predictive coding is a gamechanging technology and its almost riskfree.

925 143 214 1027 687 1168 1273 173 1485 1073 1318 631 369 1428 78 346 1026 896 1309 543 1105 343 530 301 112 521 790 1298 1211 1498 1389 275 401 1013 489 160 1127 1096 630 341 24 255 1413 46 265