Product Feature Spotlight: Deep Coding

18 August 2022 | Luminance

Over the past decade, eDiscovery has become integral to the litigation process. With the explosion of enterprise data, nearly every investigation now includes a substantial digital evidence component – whether it be contracts, emails, text messages or GPS data. Indeed, the technologies and process surrounding Discovery have become unrecognisable to what they were in the early 2000s.

Now, Luminance, the most advanced AI for legal process automation, is allowing lawyers to gain rapid and comprehensive insight into their datasets, offering a uniquely fast and robust way for organisations to approach all aspects of document review, investigation and eDiscovery.

AI-Powered Technology Assisted Review

Luminance’s ‘Deep Coding’ is a powerful application of machine learning to the document review process, whereby the system actively learns from the lawyer’s expertise when identifying which documents are relevant to their investigation. By observing the reviewer’s coding behaviours, Luminance forms an understanding of why documents are marked as relevant or not.

With just one click, users can then choose to extend Luminance’s understanding of relevancy across the entire dataset and expedite the review through auto-coding. What this means is that Luminance will analyse the remaining dataset, plotting on a graph how relevant it anticipates the remaining documents to be. Importantly, lawyers can remain fully in-control of the legal process by setting inclusivity thresholds, which can be modified at any stage of the review.

As well as predicting document relevancy, Luminance highlights discrepancies between coding decisions on documents, including those which it believes are relevant despite having been coded as irrelevant. Crucially, this enables lawyers to check the coding behaviour of other reviewers and ensure that no important documents have been mis-coded, making Luminance an indispensable auditing tool when multiple users are reviewing a set of documents.

Conceptual Understanding and Visualisation of Documents

Another powerful feature of Luminance’s Deep Coding is the Heat Map, which provides an intelligent, visual display of the entire dataset, allowing lawyers to gain an instant overview of their review progress. With no prior training needed, Luminance’s AI organises the dataset by conceptual similarity. Luminance then displays this analysis in the form of the interactive Heat Map, with each tile representing a conceptually similar cluster of documents. The depth of a tile’s colour illustrates how many of these documents have been reviewed, ensuring reviewers can allocate resources to under-investigated areas, so that no pocket of data or potential ‘smoking gun’ is overlooked.

All of this analysis is possible thanks to Luminance’s unique ability to conceptually understand information. With Luminance expediting the investigation process, lawyers have more time to develop strategy and offer their clients legal counsel based on a comprehensive understanding of the entire body of evidence. As Gavin Williams, Partner at Holland & Knight, has commented: “Luminance allows you to spend more time being a lawyer as opposed to a detective”.

Interested in finding out more about how Luminance’s AI-powered eDiscovery tool is placing lawyers on the front foot prior to pre-trial negotiations and court hearings? To request a demo, click here.