Discovering electronically stored information has never been exclusive to litigation. Volume, scope and privacy concerns have increased the complexity of all investigation and review processes. Discover how process rigor and enabling technologies ease the burden.
Data Privacy
Discipline
Hoping for the best, ready for anything.
The risk of litigation looms over every investigation—internal, compliance, regulatory, breach, or due diligence. Following a well-defined workflow from the start creates the artifacts necessary to demonstrate defensible approach to data identification, collection and review. We have adapted the processes and tools we use in discovery to support investigations.
• Scale to meet the size and speed of an accelerated process
• Documented, consistent processes stand up to scrutiny
• Investigation templates and workflows in Relativity for a quick start
• Managing rolling ESI through all processing, search and review stages
Knowledge Work
Enabling technology to accelerate learning
Investigations often include broad collections, which means large data sets. Discovery tools make smart work of surveying your data to know where to start digging. We combine Relativity Analytics, NexLP or Brainspace with our technical experts to help you survey content. We are here to explain the technology, teach users, look over your shoulder or fully support your team with linguists and data scientists.
• Sampling strategies to uncover unique and unanticipated content
• Analytics, including categorization, clustering, conceptual searching
• Threading and communication analysis
• Technology Assisted Review
Sensitive Data
PII is just one type of sensitive data at risk.
It may not be the focus of your investigation, but sensitive data is certainly caught up in the data. In a data breach meeting noticing obligations is an initial focus, but understanding strategic, IP, operational and simply embarrassing data exposed is important to protecting your assets and managing your public image. Similarly, responding to a DSAR means protecting all the non-subject data contained in systems and documents. We use all of our discovery tools to identify and target sensitive data.
• Categorization and conceptual searching
• Regular expressions to identify known data formats
• Sentiment analysis to identify sensitive content
• Image classification with computer vision machine learning
• DSAR solution to limit the exposure of all non-subject data
Private Information Identification – with precision
Problem
Traditional approaches to identifying protected private information are often overbroad, and many do not capture a wide range of sensitive health information. These limitations stress already tight timelines for review and redaction.
Solution
ProSearch deployed Privacy Suite—a combination of deep learning and visual document classification models to identify private information and to classify medical and health information at a sentence level. These techniques are optimized to be compatible with Active Learning workflows.
Result
With private information classified at a granular level, review queues can be designed to target specific business needs. Solutions can accommodate redaction or withholding in a production, can inform reputational risk assessments, and can support breach notification workflows – including varying and evolving jurisdictional-specific requirements.