Everlaw, the cloud-native investigation and litigation platform, unveiled its Clustering software feature today, delivering an AI breakthrough in terms of its scale, visualization, ease of use and ability to conduct true discovery.
While Technology Assisted Review (TAR) has been sanctioned for legal teams to conduct discovery searches for digital evidence for about a decade, the promise of concept clustering has fallen short. It’s often too hard to use, can’t scale to meet today’s video, audio and text demands, and is restricted to a wheel interface that can’t drill down to single documents.
Everlaw Clustering’s new technical breakthroughs deliver on the promise of AI, allowing legal teams to sort through and understand millions of documents for full review or early case assessment (ECA). Everlaw Clustering presents findings in an intuitive visual format that encompasses both a 30,000-foot snapshot and a granular, down-to-the-document view. It uses unsupervised machine learning to group documents by conceptual similarity and generates insights without requiring any user input. With a clean and easy-to-use interface, review teams instantly gain a baseline understanding of the document set without advanced setup or deep technical expertise. It is designed to pinpoint more specific and relevant information than other AI tools or keyword searches, and quickly identify which documents need human review, reducing the risk of errors in ediscovery.
Everlaw Clustering breaks new ground in these areas:
Extreme Ease-of-Use – Based on patent-pending technology, Everlaw Clustering was designed to be as easy to use as navigating Google Earth. In a breakthrough of visual display (see video here), the corpus of data can be seen in its interrelated relationships at a high-level view with specific clusters clearly labeled and color coded. Users can quickly zoom into individual documents.
“Everlaw Clustering allows legal teams to move from a planetary survey of all the evidence in a case all the way down to a blade-of-grass view – a single document – in a unified view,” said Everlaw CEO and Founder AJ Shankar. “Spatial relationships are preserved between documents and clusters, enabling teams to intuitively explore related concepts and discover – in the true sense of the word – new evidence. It is among our biggest achievements to date for its ability to deliver cutting-edge analytics in a consumer friendly design, at a scale relevant to today’s most sophisticated cases.”
Massive Scale– Everlaw Clustering’s unsupervised learning algorithms scale to new heights in ediscovery, supporting up to 25 million documents on its single screen clustering dashboard in a format that is unique to the data itself and breaks from the traditional wheel visualization. Everlaw Clustering’s scalability is particularly useful in the ECA stage, when teams face evidence that may number in the millions and consist of various emerging data types.
“Everlaw Clustering has reinvented the wheel for ediscovery,” said Shankar.
True Discovery – Everlaw’s spatial model preserves relationships between documents, even across clusters and zoom levels. It enables a natural, fluid exploration that allows teams to truly discover new evidence as they map out their haystacks and then build compelling storylines.
Clustering’s rich review capabilities incorporate overlaying ratings, codes and predictive coding models. For example, in the review phase of ediscovery, users can utilize existing predictive coding models by overlaying prediction scores onto the visualization to identify clusters that contain many predicted hot documents – pointing them to additional potentially relevant documents in those clusters. Clustering’s overlays can also help conduct quality checks by calling out anomalous ratings and codes to be assured that documents are reviewed consistently, further removing risks of human error.
“Everlaw’s approach demonstrates the newest AI techniques. It not only looks different, but has an intentional design to move from a linear view to a functional relationship of the data in an intuitive and cognitive manner,” said Ryan O’Leary, Research Manager, Privacy and Legal Technology, IDC. “The scale of its data consumption has the potential to raise the bar for today’s ediscovery capabilities.”
A Platform for Integrated, Advanced AI
Everlaw Clustering is seamlessly integrated with the Everlaw platform to help legal teams accelerate finding key pieces of evidence, mitigate the risk of human error and confidently navigate ediscovery at terabyte scale. Clustering also complements Everlaw Predictive Coding’s supervised learning for more powerful AI workflows. Everlaw Clustering is included in the Everlaw platform.
More specifically, Everlaw Clustering enables legal teams to:
-
See clusters dynamically separate and merge based on zoom level through dynamic zoom
-
Overlay predictive coding models and use the prediction scores to find hot documents
-
Overlay ratings and codes to prioritize certain document sets or validate review decisions
-
Recluster at any given moment
-
View the most common terms found in clusters and any transcribed A/V files
-
Filter visualization to only display a specific search
-
Access similar documents in a cluster through the context panel in document review window, and
-
Open documents directly into Data Visualizer