Axcelerate eDiscovery Results
Efficient and Effective Review
Predictive Coding in Action
What is Predictive Coding?
Predictive Coding is a court-endorsed process that combines people, technology and workflow to find key documents quickly, irrespective of keyword. Due to its massive accuracy and efficiency gains, Predictive Coding is revolutionizing how Early Case Assessment (ECA), analysis and document review are done. Predictive Coding has three components:
Case experts use Predictive Analytics to find key documents quickly and irrespective of keywords
Keyword-agnostic machine learning finds other relevant documents
Proven workflow with integrated sampling delivers results to a statistical certainty
Recommind developed Predictive Coding in partnership with some of the world’s leading enterprises and law firms. Recommind customers have been using Predictive Coding for the past 5 years.
What does Predictive Coding require?
Predictive Coding requires ALL of the following:
- Input from a case expert
- Keyword-agnostic analytics to find key documents and create seed sets
- Proven workflow to deliver statistically certain results
- Iterative machine-learning to “find like this” based on meaning not keywords
- Integrated sampling for certainty and unparalleled defensibility
What does Predictive Coding not do?
Predictive Coding does not replace human review. It optimizes it. The solution takes all the documents related to an issue, ranks and tags them so that a human reviewer can look over the documents to confirm relevance. The beauty of this technology is that attorneys can use human decisions to teach the computer, making the relevancy suggestions more accurate over time.
What is Technology Assisted Review?
Technology Assisted Review (TAR), also known as Computer Assisted Review, is not Predictive Coding. Technology assisted review includes aspects of the non-linear review process such as culling, clustering and de-duplication, but TAR does not meet the requirements for comprehensive Predictive Coding as outlined above.
Many technologies assist in making incremental reductions in eDiscovery costs, but only fully integrated Predictive Coding is able to completely transform the economics of eDiscovery.
Patented Predictive Coding Technology
The proprietary technology and deep knowledge behind Recommind’s Axcelerate eDiscovery differentiates it from late comers in the market. The Predictive Coding patent and additional patents pending give Recommind, its customers and its partners exclusive rights to use, host and sell systems and processes for iterative, computer-assisted document review.
Traditional eDiscovery software relies on linear review, a tedious, expensive and error-prone process in which teams of contract attorneys review hundreds of thousands of documents one at a time. With a small amount of information from a knowledgeable attorney, Predictive Coding uses machine learning to categorize and prioritize any document set faster, more accurately and more defensibly than contract attorneys, no matter how much data is involved.
By giving human reviewers computerized assistance, Predictive Coding saves 60 to 90 percent on review costs – savings that can easily reach millions of dollars per case or across several small cases.
In addition, only Recommind has the patented CORE® technology that automatically understands the latent concepts within an entire document collection. Predictive Coding’s use of this sophisticated technology provides a much more thorough and accurate review, providing the highest degree of accelerated cost and time savings in the market.
To learn more about Recommind’s CORE technology, read the whitepaper Finding Information: Intelligent Retrieval and Categorization.
How Predictive Coding Works
Axcelerate’s patented Predictive Coding process is based on Recommind’s unique probabilistic latent semantic analysis (“PLSA”) technology and includes a defensible workflow to conduct automated review and coding. Predictive Coding makes a computer-generated judgment - with explicit confidence score - about the relevance, responsiveness, and/or privileged nature of each document. This capability allows outside counsel and their clients to dramatically expedite the actual review process while concurrently improving accuracy and lowering the risk of missing key documents.
Predictive Coding: The Next Phase of Electronic Discovery Process Automation
Produced by the Enterprise Strategy Group (ESG), this white paper addresses the necessity of predictive coding in the context of broader trends in legal sector automation and outlines considerations for comparing and evaluating different approaches in the market.
Read the Predictive Coding white paper here
Predictive Coding Explained
Authored by Recommind’s Chief Technology Officer, Dr. Jan Puzicha, this white paper focuses on the technology and workflow behind this groundbreaking document review process, pioneered and patented by Recommind.
Read the Predictive Coding white paper here
WilmerHale Dramatically Upgrades Review Capabilities, Ensures Quality Control with Axcelerate Review & Analysis
After careful consideration and thorough product trials, WilmerHale selected Recommind as its technology partner to develop an innovative approach to eDiscovery. In this case study, learn how Wilmer uses advanced analytics and non-linear review to provide its clients with better, quicker matter insight and more predictable document review costs.
Read the WilmerHale Predictive Coding case study here
Axcelerate Review & Analysis Reduces Time and Expense for Fulbright & Jaworski
Learn how Fulbright & Jaworski was able to increase first pass review quality and reduce the length and expense of review by leveraging Axcelerate eDiscovery’s Predictive Coding.
Read the First Pass Review case study here
Eimer Stahl Gains Faster, More Consistent Document Review with Recommind
In this case study, read how Eimer Stahl, a boutique law firm specializing in complex commercial litigation, used Axcelerate Review & Analysis with Predictive Coding to reduce resources and overall cost of its document review.
Read the Eimer Stahl Predictive Coding case study here
The Defensibility of Non-Human Document Review
In this BNA article, WilmerHale attorneys Robert W. Trenchard and Steven Berrent explain how combining advanced search with statistical testing results in a process that can satisfy the Federal Rules’ requirements for a ‘‘reasonable’’ review.
Read the eDiscovery Defensibility article here
How Fulbright & Jaworski Transformed Its Document Review
In this Legal Tech Newsletter® case study, read how Fulbright’s strategic deployment of Axcelerate eDiscovery, and particularly its Predictive Coding technology, revolutionized the firm’s document review process.
Read the Predictive Coding case study here
Recent blog posts
April 15. Unless you are a tax lawyer this day likely is not one of your better days. Even if you are a tax lawyer, you know that today will be the longest day of the year (next to the last day of the firm’s fiscal year as you await that check the client promised [...]
April 1 has come and gone, and the legal profession remains eager with anticipation to see what the overall impact will be from the recently enacted Civil Justice Reforms in England and Wales. The reforms, often referred to as the Jackson Reforms based on the January 2010 report of Lord Justice Jackson which served as [...]
It’s that time of year. Law offices across the country have been slightly understaffed the past few weeks, but for most of us Spring Break 2013 is now a memory. Beautiful blue skies, nice weather, and free flowing libations at a scenic resort in southern California; yep, San Diego this time of year is the [...]
A key phase in eDiscovery is Early Case Assessment (ECA), the process of reviewing case data and evidence to estimate risk, cost and time requirements, and to set the appropriate go-forward strategy to prosecute or defend a legal case – should you fight the case or settle as soon as possible. Early case assessment can [...]
Predictive Coding Training Program
The goal of the program is to provide a clear understanding of what Predictive Coding is, how it works, how it is employed and how attorneys and litigation support staff can get started using it. The program offers deliveries proficiency training for three levels: Predictive Coding Reviewers, Predictive Coding Strategists and Predictive Coding Sys Admins.
In light of the ground-breaking opinion in DaSilva Moore v. Publicis approving the use of Predictive Coding, Recommind is offering a three-tiered educational campaign on Predictive Coding for Reviewers, Strategists, and Sys Admins.
The goal of the program is to provide a clear understanding of what Predictive Coding is, how it works, how it is employed and how attorneys and litigation support staff can get started using it.
Training: Gain the knowledge and skills to maximize efficient and productive use of Axcelerate Review process including:
- Predictive Coding workflow
- Optimal use of all Axcelerate tools in the workbench: Filters, review choices, document view options, redaction
- Best practices for templates, searches, advanced searches
Seminar: “Predictive Coding: The New Economics of Document Review”
Attend a seminar to hear from an experienced professional on Predictive Coding and how it is changing document review in the legal industry. These seminars are open to: Corporate counsel, assistant general counsel, litigation counsel, law firm litigation partners, senior associates, eDiscovery team members, and paralegals.
Strategist Training: Gain the knowledge and skills to oversee the whole review or investigation process including:
- Knowledge and skills to manage all steps and phases of the PC workflow
- Knowledge of best practices including:
- Advanced workflow
- What/how to effectively use sampling
- Management of adding data
Seminars are open to Senior Attorneys.
Strategist Training is open to Recommind Customers and Partners only.
Sys Amin Training: Gain the knowledge and skills to configure, maintain and enhance the technical environment for Predictive Coding workflow, manage and perform ingestion of data stores, productions, log analysis, server architecture best practices, configuration files and connectors.
Sys Admin Training open to Recommind Customers and Partners only.
Traditional linear review is a manual, expensive, time-consuming and error-prone process, requiring teams of legal professionals to review hundreds of thousands or millions of documents one page at a time to determine relevance to the case.
On the other hand, Predictive Coding utilizes iterative machine learning to produce a computer-generated judgment with an explicit confidence score about the relevance of each document. This capability allows counsel and their clients to dramatically expedite the actual document review process by prioritizing each document while concurrently improving accuracy and lowering the risk of missing key documents.
The savings calculator below first calculates the projected cost of review using a traditional linear review strategy. Then a projected cost is calculated using Recommind’s Predictive Coding application. Enter in the approximate number of GB of data for your case, adjust the approximate number of documents per GB if needed and hit “Calculate” to see your estimated savings:
The end result with Predictive Coding is human reviewers actually read a much smaller percentage of the corpus. In most cases you will see a 75% to 95% cost and time savings when relying on Predictive Coding for document review over that of traditional linear review.