Even before the exponential increase in digital data production, legal professionals faced tight deadlines as they navigated through the paper material to find key evidence. Today, legal professionals are even more burdened as they navigate through an overwhelming number of e-mails, text messages, personal computers, shared drives, social media data, and more to find key evidence. Legal professionals not only have to review these documents in a timely manner, but also have to ensure accuracy and cost efficiency.
Knovos’ eZAnalytics™ software supports a Technology Assisted Review (TAR) module that provides efficient methodologies for case teams to prioritize, categorize, or qualify documents during the review process. The TAR solution can be used as a predictive coding tool or as a Continuous Active Learning (CAL) tool. Built into Knovos’ document review solution, the TAR module is completely integrated with the case review strategy.
The predictive coding methodology is used once subject matter experts perform a human review on a small sample set of the documents. Using human coding decisions and computer-generated algorithms, the machine then undergoes an iterative training process until it achieves the desired precision and recall levels. The machine then codes the rest of the documents in the data set, drastically reducing the need for costly human review.
Our CAL methodology provides review teams with another way of implementing an efficient review strategy. CAL allows review teams to pull in what the machine considers “top-ranked” documents to prioritize the review, drastically saving time by directing documents that are more likely to be relevant to the top of the heap. As the review progresses, the machine continuously learns from the coding decisions of reviewers, increasingly becoming better able to differentiate between responsive and non-responsive documents.
Both of these methods streamline the document review process by reducing the need for human reviewers and by enabling case teams to establish efficient case strategies — before and during the review. In addition, our TAR solution offers quality control capabilities, including random sampling and conflicting decision reports that quickly identify inconsistencies in decisions between different document groups. With our TAR solution, efficiency is accompanied by consistency and accuracy, ensuring case teams of the quality of their work product.