Technology Resources for Arbitration Practitioners - Document collection, review and production
Technology Resources for Arbitration Practitioners
Document collection, review and production
The collection, review and production of electronic documents can be one of the most costly and time-consuming phases of arbitration, particularly where there is a vast number of potentially relevant electronic documents to review. Advances in software engineering and machine learning may render the processes of collection, review and production of e-documents more efficient and cost effective.
The technologies presented in this chapter are organised into the following three categories: document review platforms, software/AI applications for document review and specialised vendors for document review.
The use of document review software platforms has become very common in arbitrations involving a large number of documents. This software allows large volumes of data to be uploaded, reviewed, coded and produced from a single platform that can be hosted in a single central location (ie, at a law firm, client or vendor premises) or remotely through a secure connection accessible over the internet. Once the documents are hosted on a software platform, there is potential to reduce the volume of electronic data through early document culling tools, such as deduplication, near duplicate identification and email threading. Documents can be separated into more manageable categories by converting image files through optical character recognition (OCR) into editable and searchable documents, indexing documents, key term identification, concept clustering and foreign language identification. Software programs can also identify media files, transcribe their contents and search within them to help locate key documents.
Some examples of document review platforms are below:
Software platform designed for the processing, review, analysis and production of electronic documents It can host large amounts of data and is accessible through the internet. Bringing all data together in a single platform, Relativity offers advanced searching tools, machine learning and applications built from over 50 developers to assist users with the review and production of documents.
The data must first be gathered and then is uploaded onto the Relativity platform. Various vendors are familiar with Relativity and can be hired to assist with the gathering, uploading, review, hosting and production of data. Once transferred into Relativity, data can be accessed through a hosting partner, or through internet or cloud connections.
Relativity includes processing capabilities, computer-assisted review, image and native file review, searching, diverse coding options, flexible workflow capabilities, integrated productions, foreign language support and text analytics, among other features.
Document review and production platform with a visual approach that can run on the cloud or through a hosting partner The software features multi-dimensional early case assessment; visual analysis of concepts, custodians and critical data points; visual predictive coding, pivot tables, Boolean searching, keyword searching, concept clustering, timelines and email threading, among other features.
There are myriad software applications that can be used within document review platforms to further assist with all stages of document collection, review and production.
- At the collection stage, for example, specialised software can assist with the collection and review of documents in unstructured data formats (eg, chats, Multimedia Messaging Service (MMS) and Short Message Service (SMS) messages) by grouping the messages into a coherent format that the reviewer can read sequentially.
- Software can help reviewers to visualise data in the documents that would not be easily apparent from a linear review. For example, there is software that allows users to quickly access trends in data types and volumes through a bird’s-eye view of the flow and volume of email, or email timelines. Another example is social network visualisation, which simplifies the document view by consolidating email aliases into a single identifier representing each person. This helps to identify relevant information as well as map the participants in relevant conversations.
- At the production stage, auto-redaction software tools can reduce the time needed for the linear review of documents containing privileged and sensitive information. There are also tools that allow the markup and redaction of voluminous native Excel and image files. Redactions applied in one document can be automatically applied to duplicate content in other documents. There are also software applications that can automatically find and redact sensitive information, such as sensitive words, phrases, social security numbers and credit card numbers.
- Parties may choose to apply some form of technology assisted review (TAR), continuous active learning, and/or predictive coding to categorise and code massive amounts of documents in a potentially shorter period of time. Continuous active learning allows users to train a model to identify responsive documents through the review and modelling of a small initial set of documents. The software then applies the model to the remaining set of documents and provides suggestions for the coding of those documents. As the reviewer conducts a secondary review of the coded documents, the software adapts and changes its coding to match the inputs from the reviewer.
The following is a non-exhaustive list of software developers who make some of the applications discussed above:
There are hundreds of vendors that can assist with the collection and processing of electronic data, host data for large-scale document review, assist with and manage document review, and handle document production.
Many of these vendors use the Relativity software platform and some offer their own document review software platform as well as a number of other platforms from which parties can choose.
The following are a few examples of vendors, which often vary in popularity and experience depending on the region in which they operate:
Offers document collection, review and production services, including Relativity software hosting It has developed its own suite of technological tools to assist with review, such as its own review platform (Sightline), enhanced audio review and enhanced chat review.
Hosts data and offers review platforms, such as Relativity, as well as its own platform DocuMatrix (DMX). DMX software is a processing, review, production and document review management platform. Epiq also offers Arq, which is a Relativity environment for a fixed monthly fee, allowing access to Relativity without the large initial expenditure of hardware and licensing costs.
Provides technology-enabled software and services to law firms from forensic data collection through to document production. It offers Nebula, which is an end-to-end document review software that allows users to cull, process, review and manage documents, and is hosted on the cloud. Nebula includes predictive coding software; native spreadsheet redaction; A/V Suite, which simplifies the review of multimedia files and allows redaction of audio files; auto redaction; automated privilege logging; and automated workflow. KLDiscovery also offers Relativity hosting, and many of its applications can be used in Relativity as well as Nebula. In addition, KLDiscovery offers its own selection of software applications, such as Ediscovery.com Review, which is used for conducting early case assessment, analysis, review (including predictive coding) and document production; and KLD Analytics, which offers a full range of technology-assisted review tools.
Based in Australia and Asia, offering data hosting, document collection and review services, as well as its own suite of proprietary software
Offers data hosting, document collection and review services, as well as its own suite of proprietary software, such as BlackBar (auto redaction software for native excel files).
In addition to the above vendors, many large accounting firms offer data hosting, document collection, and document review services, including but not limited to:
- Ernst & Young
In addition, several consulting firms offer similar services, including but not limited to:
- FTI Consulting
- Control Risks
- Alvarez & Marsal
Because the list of potential vendors in this area is so large, we have included below a link to a source that lists various document collection providers:
Disclaimer: Due to the very nature and dynamics of the subject of this guide, the examples should not be considered exhaustive, and merely represent a sample of the potential applications available. There are numerous other vendors that provide similar services and products to the ones described, and the presence of any particular vendor or product in this guide does not reflect any qualitative judgment about the suitability or capability of that vendor or product. The goal is to periodically update and edit the guide to reflect new technological advances, and add new or delete obsolete, applications, programs or vendors. The IBA Arb40 Subcommittee does not endorse or recommend any particular technology, vendor, software or program listed below, nor can it vouch for the security, cost or appropriateness of any of the listed technology, which must be assessed by practitioners on a case-by-case basis. The descriptions of particular programs, software and vendors were not provided by the vendors themselves, and the IBA Arb40 Subcommittee takes no responsibility for errors in those descriptions. All technology should be thoroughly explored and vetted by the arbitration practitioner prior to use.