P2P

11 I L T A N E T . O R G W ith the deregulation of the practice of law and advancements in technology, the legal sector has undergone enormous changes. To better deal with growing volumes of data in the age of digital transformation and tight timelines in high-stakes litigations, the field of discovery has seen an accelerated adoption and application of data analytics. In particular, the rise of data analytics has helped improve the quality and speed of document review. However, the benefits of using a scientific and analytical approach to data are not limited to discovery-related processes. Designed to reveal trends and promote efficiency, the data analytics commonly used in discovery can help law departments organize, analyze and understand their business data as it grows in volume and complexity. Crucially, questions related to the business needs of law departments are not dissimilar to the questions that serve the needs of discovery. As law departments seek to do more with less, insights into how they operate and deliver services, derived from the data they already own, can help lead to reduced costs and improved operational efficiency. These insights and the fact-finding missions that uncover them are not unlike the insights from discovery investigations performed for litigation and compliance purposes. Essentially, they address the questions of what, how and why problems occur and who causes those problems, regardless of whether these questions are being asked to improve the business outcomes of law departments or to gather evidence for a court of law. Given the massive and complex nature of law departments' data, data analytics in discovery enables law departments to separate the wheat from the chaff and make sound business decisions. There are three main types of data analytics in discovery: structured analytics, conceptual analytics and predictive analytics. The first type, structured analytics, refers to a range of different techniques that analyze the textual similarities and differences between documents within a set. The most common application of structured analytics is to help reduce the amount of duplicative content in a dataset. Discovery platforms generally impose at least some form of exact deduplication by default to minimize repetitive material, but textually identical documents are not the only source of duplicative content in a dataset. If a user does not want to consider all of the documents that have only minor differences from one another, near-duplicate detection can be used to identify a set that is above a selected minimum percentage of similarity. Then, the user can save a significant amount of time by looking at only one document within each near-duplicate group. Another powerful technique used to drive down the proportion of duplicative content in structured analytics is email threading. As emails are exchanged between the sender and recipient(s), the original content is typically included by default in a reply or a forwarded message. Therefore, duplicative content is generated because the same information can now be found in more than one "The rise of data analytics has helped improve the quality and speed of document review."

• Cover