Computerized categorization systems compare favorably to manual review in conducting e-discovery
A study by the Electronic Discovery Institute has demonstrated that automated review is a credible alternative to manual document review for managing large volumes of electronically stored information (ESI) in a time-saving and cost-effective manner. The study evaluated automated technology from two independent providers including
Cataphora Inc., an industry-leading developer of software for review and analysis of electronic data.
The study, titled Document Categorization in Legal Electronic Discovery: Computer Classification vs. Manual Review, compared computer classification of electronic documents to traditional manual review in order to determine whether or not the automated review systems could serve as an acceptable alternative to human review teams. The study used an original manual classification, already conducted in response to a Department of Justice Request, as a standard for comparison against further categorization undertaken by two human review teams and two computer systems, one of which was Cataphora’s. To gauge the classification accuracy of the human reviewers and computerized systems, the decisions made by the human review teams and the computer systems were compared to those made by the original review team to see how well they aligned. The results showed that automated review is as effective in categorizing documents as manual review, if not more so, affirming computer categorization as an invaluable tool for conducting effective document review.
“The results of this study clearly show that Cataphora’s automated review offerings represent a highly desirable option for organizations seeking to conduct timely and defensible document review in the face of swelling costs,” commented Cataphora’s chief linguist, Dick Oehrle, Ph.D. The study illustrates how the use of computerized systems can tackle large quantities of ESI while helping attorneys satisfy Rule 1 of the Federal Rules of Civil Procedure: “to secure the just, speedy, and inexpensive determination of every action and proceeding.” From its inception in 2002, Cataphora’s mission has been the development of practical leading-edge technologies such as automated review. The company has proven the advantage of automated review as a mechanism for cutting the cost and time of the e-discovery process to a multitude of satisfied customers since the company first introduced it in 2004.
The study was authored by Herbert L. Roitblat, Anne Kershaw, and Patrick Oot, all from the Electronic Discovery Institute (www.ediscoveryinstitute.org). The study used 1.3 terabytes of data, consisting of 2,319,346 documents, which were obtained and processed by the original review team in response to a “Second Request” regarding the acquisition of MCI by Verizon. The human reviewers conducting the original categorization worked 16 hours a day, seven days a week, for about four months. By contrast, computer systems can operate 24 hours a day without pause, rest or distraction. In the end, the decisions made by the computer systems coincided with those made by the original reviewers at least as often as did those of the human review teams in the study.
The Electronic Discovery Institute is a non-profit organization dedicated to resolving electronic discovery challenges by conducting studies of litigation processes that incorporate modern technologies. The Institute operates under the guidance of an independent advisory board comprised of judges, lawyers and technical experts. The Institute’s studies measure the relative merits of new discovery technologies and methods.
About Cataphora:
Cataphora Inc. is the creator of core patented technology that provides clear, accurate insight into individual and organizational behavior. Cataphora products – based on advanced analysis of multiple sources of electronic data – can be used in situations such as litigation, investigation, workforce restructuring, post-merger integration, and enterprise governance, risk management, and compliance (GRC). Cataphora’s unique technology constructs complex models of organizational behavior, allowing any deviations from normal behavior to be more easily identified. It is the first information retrieval technology to recognize context that spans different records and even different record types. It does so by identifying causal relationships among these various records and creating a representation that supports superior analysis of the approximately 80 percent of corporate data that is unstructured.
Founded in 2002, the company is headquartered in Redwood City, Ca., with offices in Washington D.C., and Frankfurt, Germany, and a field office in France. For more information, please visit www.cataphora.com.
