Technology Assisted Review with Predictive Coding

Document Decisioning Suite for Technology Assisted Review ("TAR eDiscovery")

Now with Redaction!

Click here to arrange a demonstration.

Studies show that people are only 20% effective at guessing the correct search words to use in finding information. The OrcaTec™ Document Decisioning Suite deploys technology assisted review that enables “learning” the meaning of words from their context in the electronic documents. The Suite is able to examine both the actual words and their meaning – thus, it can search for the “concept” or “topic”.

Independent validation

Independent study¹ by Maura R. Grossman & Gordon V. Cormack validates technology assisted review:
jolt.richmond.edu/v17i3/article11.pdf
Please read this article. It will change the way you think about eDiscovery and automated legal document review!
 

Search

▶ Patent-pending, visual Concept Search reduces the burden of guessing the right words to search. It learns automatically what words mean throughout the information set, without the labor and expense of dictionaries, ontologies or thesauri.

▶ Search Suggestions automatically return a selectable list of words and phrases from the collection (including alternative spellings), significantly improving the “find ability” of documents.

▶ Near Duplicate Identification brings together documents that are nearly identical, enabling users to compare and make faster, more consistent document decisions.

▶Email Threading organizes emails into conversations, revealing the context of the communication and reducing review time by 50% or more.

Categorize

▶ Advanced Predictive Coding is also called categorization or automated review.

▶ It essentially looks over the shoulder of an expert, capturing and replicating their decision patterns, and identifying which documents in the set contain relevant concepts.

▶ This automated process requires the least amount of manual effort possible, thus dramatically reducing the cost of review.

▶ First Pass Review of huge document sets can be completed at a rate more than a thousand times faster than human review alone, resulting in a focused set of highly relevant documents for final review that can be culled through in hours rather than days or months.

Review

▶ Review Tags can be customized for the needs of each document collection, including tagging priority, which tag types override others, and to whom documents with a particular tag should be assigned for additional review.

▶ OrcaTec Review’s interface is directly accessible from any other section of the Document Decisioning Suite, so documents can be tagged on the spot as they are uncovered, allowing for bulk tagging of attachments, conversation threads and near duplicates.

▶ All tagging decisions are fully auditable, where the performance of an individual review can be tracked down to the most minute detail.

▶ Bulk tagging permits rapid and immediate assignment of tens or thousands of documents at a time.
 

Cluster

▶ Patent-pending Cluster technology automatically identifies conceptually similar documents and groups them visually, allowing document review by topic. Cluster relevance can then be determined by examining only one or a few of the documents in the cluster, then the whole group of documents can be tagged at once.

▶ Cluster labels are used to identify key terms.

▶ OrcaTec Cluster allows the same document to be in more than one cluster, so the user doesn't have to worry about throwing away a responsive document because it also contains non-responsive information. For example, if an email is about two things, “bringing pizza to the party on Saturday” and “putting the stolen money in a Swiss bank account”, there is no danger that the email will end up only in the “pizza party” cluster where reviewers will not see it again. The email will reside in both the “pizza party” and the “Swiss bank” clusters.

▶ While reviewing the clusters, it is easy to review some or all of the documents, as well as examine time lines and social networks to see who communicated with whom, when and about what. By being able to focus on relevant documents quickly, the investigator can more efficiently and effectively get to the heart of the matter.
 

Clusters automatically group documents visually by topic

Timeline provides a temporal view of documents and concepts

Redaction

Sonar allows zoom-in capability on topics, branches and nodes

Network displays communication paths and density

Please contact us for more information, or click here to register and to get access to the product demonstration.

1. Maura R. Grossman & Gordon V. Cormack,
Technology-Assisted Review in E-Discovery Can Be More Effective and More Efficient Than Exhaustive Manual Review, XVII RICH. J.L. & TECH. 11 (2011), jolt.richmond.edu/v17i3/article11.pdf.