Technology-Assisted Review in E-Discovery

What is Technology-Assisted Review?

Technology-Assisted Review (TAR), also known as predictive coding or computer-assisted review, is a process used in e-Discovery to expedite and improve the efficiency of document review. It involves the use of machine learning algorithms and other advanced technologies to assist in the identification and prioritization of relevant documents for legal review.

Traditionally, e-Discovery involves the manual review of large volumes of documents to determine their relevance and responsiveness to a particular legal matter. This process can be time-consuming, costly, and prone to human error. TAR aims to address these challenges by leveraging technology to automate and streamline the review process.

Here's an overview of how TAR works:

1. Seed Set Creation: The TAR process begins with the creation of a "seed set" of documents that are manually reviewed and categorized by human experts. This set represents a sample of the documents in the larger collection and serves as the basis for training the machine learning algorithm.
2. Machine Learning Training: The machine learning algorithm is then trained using the seed set. The algorithm analyzes the characteristics and patterns of the reviewed documents to learn how to identify relevant documents based on various criteria, such as keywords, concepts, or document metadata.
3. Iterative Learning: Once the algorithm is trained, it is applied to the remaining documents in the collection. The algorithm assigns relevance scores to each document, indicating the likelihood of it being relevant to the legal matter. Initially, human reviewers may validate and correct the algorithm's predictions to improve its accuracy. These corrective actions are used to refine the algorithm further, creating a feedback loop of iterative learning.
4. Sampling and Quality Control: Throughout the TAR process, random samples of documents are periodically selected and manually reviewed to assess the algorithm's performance. This helps ensure that the algorithm continues to achieve the desired levels of accuracy and recall.
5. Document Prioritization: As the TAR process progresses, the algorithm ranks the documents based on their relevance scores. This ranking allows legal teams to focus their review efforts on the most relevant documents first, potentially reducing the overall review time and costs.

Regulations Around E-Discovery

The regulation around e-Discovery varies across jurisdictions, but there are some common themes and guidelines that govern the process. Here are key regulatory aspects related to e-Discovery:

1. Federal Rules of Civil Procedure (FRCP) (United States): In the United States, the Federal Rules of Civil Procedure govern e-Discovery in federal courts. These rules outline the scope, obligations, and procedures related to the discovery of electronically stored information (ESI). The FRCP encourages parties to meet and confer to address e-Discovery issues, emphasizes proportionality, and provides guidelines for the preservation, production, and spoliation of ESI.
2. International Organization for Standardization (ISO): ISO has developed standards related to e-Discovery processes. ISO 27050 provides guidelines for the management of electronic evidence, while ISO 27037 offers guidance on the identification, collection, and preservation of potential electronic evidence.
3. Data Protection and Privacy Laws: E-Discovery intersects with data protection and privacy laws, particularly in jurisdictions with comprehensive privacy regulations. For example, the General Data Protection Regulation (GDPR) in the European Union imposes strict requirements for the handling, transfer, and protection of personal data during e-Discovery. Other countries, such as Canada with the Personal Information Protection and Electronic Documents Act (PIPEDA), have similar provisions.
4. Jurisdictional Considerations: Different jurisdictions may have specific rules and regulations governing e-Discovery. For instance, some U.S. states have their own e-Discovery rules that complement or supplement the FRCP. Similarly, countries like Australia, Canada, and the United Kingdom have their own guidelines and regulations related to e-Discovery processes.
5. Case Law Precedents: Judicial decisions and case law play a significant role in shaping e-Discovery practices and guidelines. Courts' interpretations of existing laws and rules, as well as their rulings on e-Discovery-related disputes, provide guidance and establish precedents for future cases.
6. Industry Regulations: Certain industries, such as financial services, healthcare, and telecommunications, have specific regulations and guidelines that impact e-Discovery. Organizations operating in these industries must consider industry-specific regulations, such as the Health Insurance Portability and Accountability Act (HIPAA) in healthcare or the Payment Card Industry Data Security Standard (PCI DSS) in the financial sector, when conducting e-Discovery.
7. Cross-Border Considerations: Cross-border e-Discovery involving data transfers between jurisdictions requires compliance with applicable regulations and laws. In cases where personal data is involved, data transfer mechanisms, such as standard contractual clauses or binding corporate rules, may need to be employed to ensure compliance with data protection regulations.

It is essential for legal practitioners and organizations to consult legal experts familiar with the specific regulations in their jurisdiction and industry to ensure compliance with relevant e-Discovery requirements. Staying informed about regulatory developments and following best practices in e-Discovery can help mitigate legal risks and ensure efficient and defensible processes.

Bottom Line

It's important to note that while TAR can greatly enhance the efficiency of document review, it doesn't replace human involvement entirely. Human experts still play a crucial role in training the algorithm, validating its predictions, and conducting quality control checks.

TAR has gained significant popularity in the legal industry due to its ability to save time, reduce costs, and increase the consistency and accuracy of document review. However, its adoption may vary based on jurisdiction, case complexity, and the preferences of legal practitioners and courts involved in e-Discovery processes.

Research Partner- Jharna Jagtiani