As data volumes skyrocket, greater proportions of enterprise data are unstructured (it’s as much as 80 percent), and privacy laws proliferate, it’s mission critical for organizations to uncover insights quickly, focus review, and accurately identify sensitive information—and for many, that’s where Relativity comes in. From a technology standpoint, solving these challenges requires a laser focus on research and development in artificial intelligence.
Our recent acquisition of Text IQ—a top 100 AI company—brings aboard a team of best-in-class AI researchers, engineers, and developers who deeply understand and are committed to solving these challenges. Their talent is second to none: Half of the team have advanced degrees in active learning, machine learning, or computer science, and many come with previous experience at companies like Google, Facebook, Microsoft, IBM, and Amazon.
Text IQ’s innovative approach to AI and machine learning will advance capabilities in RelativityOne and Relativity Trace, and their solutions will help customers tackle more of their big data challenges with confidence.
Understanding the Text IQ Advantage
At Relativity Fest London, we shared our vision for accessible, comprehensive, and invaluable AI seamlessly integrated into RelativityOne and Relativity Trace. With Text IQ on board, we will accelerate that vision to deliver the critical AI capabilities our customers need.
Text IQ’s approach to AI goes beyond the supervised machine learning techniques typically used in the legal industry by applying a combination of techniques—including social network analysis (SNA), natural language processing (NLP), and unsupervised and semi-supervised machine learning—to create a layer of structure on unstructured data. This makes it possible to identify sensitive information at scale and with little human input.
These types of deep learning models, leveraging the latest in AI training, can scale horizontally across massive data sets, and have the ability to continuously learn and improve as new data is added. Folding them into our platform will up-level our customers’ ability to better manage their data, and understand the stories it can tell.
Get to Know Text IQ’s Product Offerings
Already, Text IQ has applied their deep AI expertise to some of the most intimidating unstructured data challenges. Text IQ will continue to offer its core products for privacy, legal, and compliance teams—as well as its unconscious bias detector, which is part of Text IQ’s AI for Good initiative.
Access to Text IQ’s suite of products will enable our customers to solve a wider array of legal and compliance challenges. For example:
Today, Text IQ applies AI solutions to solve some of corporations’, law firms’, and government agencies’ most pressing and costly challenges related to privilege review. Because organizations must cast a wide net, conducting privilege reviews can take thousands of hours of looking for sensitive needles in a haystack. The process is also prone to human error and the risk of sharing privileged or confidential information.
Text IQ’s flagship product, Priv IQ, is proven to reduce the time and costs of conducting privilege reviews by up to 75 percent, and significantly reduce associated risks.
Current users of Priv IQ back up those statements with real-world results. For example, a general counsel at a Fortune 100 chemicals company recently told us: “Using Priv IQ made us realize we were about to turn over hundreds of documents that were privileged.”
Additionally, Text IQ is on the forefront of the next wave of privacy challenges—known as Privacy 2.0, and characterized by the increasing threat of data breaches, the proliferation of unstructured data, and heavier regulatory requirements. Text IQ for Privacy enables organizations to proactively manage personal information, particularly hard-to-find sensitive information which might otherwise remain hidden in large, unstructured data sets.
Text IQ for Privacy automates the discovery of personal and sensitive information, returns accurate and relevant results, reduces the need for human review, and significantly lowers costs. RelativityOne and Relativity Server customers can opt to take advantage of this offering to address growing privacy challenges, including:
- Data categorization that will accurately identify sensitive information and eliminate manual data mapping.
- Data breach response workflows which can reduce response times by as much as 50 percent by automatically de-duplicating and associating data to entities.
- DSAR fulfillment automation, to meet aggressive request deadlines and reduce manual review by up to 75 percent.
Tackling Your Toughest Data Challenges with Text IQ
Text IQ’s underlying AI techniques can also deliver a more comprehensive understanding of the entities in your data. Over time, integrating these techniques into our platform will empower our customers to tackle their matters more efficiently, with greater insights into the stories hidden in their data.
Central to a legal or compliance team’s understanding of any data set are the people involved and the roles they play. However, extracting those details out of large amounts of unstructured information is a huge challenge. That’s why we launched name normalization in Relativity in 2018—to help our users get a clearer picture of the individuals and all their aliases in a data set through analyzing fields within email content.
Beyond individuals and aliases, Text IQ’s technology is built to extract complete social networks from unstructured data through their Socio-Linguistic Hypergraph. The software can identify relationships between individuals, how they communicate, and how they communicate differently with different groups of people. Put simply, Text IQ finds every trace of a person in a data set—providing insights that are impossible to gain through manual review, supervised learning techniques, or limited analysis of field data.
The result is a comprehensive understanding of all individuals in a data set, even when a specific alias or name isn’t referenced.
Additionally, notable to communication surveillance teams, Text IQ has deep experience building targeted classifiers to identify very specific communication patterns and behaviors. To do this, their solutions leverage contextual clues and complex linguistical patterns to understand whether a certain communication is relevant to the nuanced behavior you’re looking for.
This capability, supported by a team of 30 AI researchers, engineers, and developers with expertise in natural language processing and deep learning, will further improve Relativity Trace’s ability to detect market abuse, market manipulation, and other forms of misconduct in communications.
Helping You Stay Ahead of the Curve
Innovation is about anticipating what’s ahead and building solutions that help you get in front of emerging challenges before they become impassable. For our customers, advanced AI and cloud connectors have rapidly become essential building blocks on that path to innovation as the shape of data has grown and shifted over time. Our recent acquisitions, combined with our organic investments, further our commitment to helping our community stay ahead of this curve.
With that in mind, and on the heels of bringing VerQu on board to make the evolving data stores of today’s enterprises more accessible, our acquisition of Text IQ will further our ability to help you extract every last drop of value out of your data—with more speed and accuracy than ever.
At Relativity, our mission to help you organize data, discover the truth, and act on it demands that we empower you to not just grab and process and deliver documents, but take control of your data in a way that makes it an asset rather than a liability. Weaving Text IQ’s state-of-the-art AI platform into our solutions is a big step in that direction. Stay tuned to see what you can do with it.
Chris Brown is the chief product officer at Relativity. He leads our product and user experience teams and is responsible for the development of Relativity’s product vision, strategy, and product roadmap in collaboration with engineering.