How Analysis Impacts E-Discovery

MetaLINCS has developed content analysis core technology that is especially suited for eDiscovery

These facilities are designed to meet specific requirements for E-Discovery, such as:

• Review must be comprehensive. Unlike web search and many other search applications, users need to examine all relevant materials, not just the top few. This implies greater emphasis on recall and precision than on finding a few good results and putting them at the top of the list.
• Important evidence is frequently subtle or obscure, using few key terms and involving only a few people. Search systems judge results to have high relevance based on indicators like extensive use of key terms and items that are frequently referenced or referred to. These factors, which work well for web search and enterprise search, do not work well for finding important evidence in E-Discovery.
• No automatic relevance ranking measures have ever been found that highlight important evidence. Important evidence is as likely to be at the bottom of a search result list as at the top using any relevance technique. Relevance ranking as an approach must be replaced by alternative techniques.
• Early in the examination of content related to a matter, it is important to identify key people, time lines, key events, and terminology used by the parties. Search is not well suited to these tasks. They require overall analysis of the collection, not a focused look at specific queries and associated individual results.