Osgoode Certificate Program in e-Discovery, Records Management, Information Governance and Privacy – Day Four
The current buzz in the e-discovery industry focuses on the use of Predictive Coding and Technology Assisted Review (“TAR”) to assist with the culling, analysis and review stages of the E.D.R.M. The use of various culling techniques, predictive coding and TAR software algorithms offer their own unique challenges but if utilized properly they can greatly reduce the number of manual review hours required to complete your review as well as lower the costs for the review portion of the litigation.
Some Day Four items of note and best practices:
Sedona Canada recommends using a combination of both manual and TAR review to achieve the most accurate results.
Ensure that you document all decisions/expectations from your Meet and Confer session, your review procedures, culling techniques, key word searching and result sampling so your entire process is completely defensible.
Technology Assisted Review tips:
Search intent should be used to determine what combination of technology and process best suits the requirements.
Although “Recall” (the percentage of relevant records identified out of all relevant records in the collection) and “Precision” (the accuracy of the relevant records identified) are both important, consideration should be given at the outset as to how each measure is weighted.
Determining the “Margin of Error” and “Confidence Level” that you are comfortable using will help determine the size of the sample that you need to use to validate your results.
The decision to use an initial “Seed Set” depends on the particular Predictive Coding software used. However, when a seed set is used, some effort should be made upfront to ensure that is is representative of the majority of relevant records, in order to maximize the efficiency of the process.
Regardless of the technology used, the results should be tested to ensure that they are valid.
Course attendees left Day Four armed with an abundance of information on various culling strategies as well as the knowledge on how to leverage technology assisted review to reduce the overall review costs for their clients.