Helping to Develop More Intuitive Techniques to Enhance E-Discovery

Knowing that our clients face ever-increasing quantities of data and ever-tightening deadlines, we continuously strive to make our applications faster, more powerful and more intuitive. Already, our Catalytics suite of analytical tools helps our clients pare down review sets, find key documents, identify confidential or privileged documents, remove spam and analyze large document collections.

Looking to the future, Catalyst is deeply involved in research and development for the next-generation of e-discovery technology. To that end, Catalyst is a founding member of the Industrial Advisory Board to The Center for Intelligent Systems and Machine Learning (CISML) at the University of Tennessee, Knoxville, a think tank formed in 2010 to promote the development of cutting-edge analytics and concept-search techniques.

Designing ‘Smart’ Computer-based Systems

CISML’s pioneering program focuses on designing computer-based systems that exhibit intelligent behavior, operate autonomously and adapt to environmental changes.

Within the university, CISML serves as the hub for collaborative research involving faculty from three colleges and five departments. They include professors who are specialists in computer science, electrical engineering, statistics, psychology and nuclear engineering. Scientists from the nearby Oak Ridge National Laboratory are also affiliated with CISML.

Examples of the CISML's diverse areas of research include pattern recognition, robotics, artificial intelligence, biologically inspired cognitive architectures, bioinformatics and data mining.

Intelligent Tools for E-Discovery

As an advisory board member, Catalyst is collaborating with CISML on research into enhanced analytical and conceptual tools for e-discovery search and information retrieval. The fruits of this research will be built into our applications for the benefit of our clients.

“Volume and complexity are the two main problems that e-discovery professionals encounter with electronically stored information,” notes CISML's associate director, Dr. Michael W. Berry. “By combining mathematics and computer science, we can find ways to make it easier for them to gather the evidence they need and make decisions about that evidence.”

A central focus of Dr. Berry’s research is clustering. Clustering is a way to take a body of data about which you know nothing and give it a structure, Dr. Berry explains. From that structure, you can identify patterns and relationships that help you decide which data is likely to be more useful and where to focus your time.

Industrial Advisory Board

CISML sought Catalyst's participation on the Advisory Board because of the company's leadership in furthering search technologies and infrastructure, Prof. Berry says. He and Catalyst previously collaborated on research into the use of nonnegative matrix factorization for document clustering and classification. That research produced joint conference papers and a book chapter on the origins and future of e-discovery search.

Catalyst is represented on the board by Dr. Jeremy Pickens, a member of the Catalyst research and development team. Dr. Pickens is one of the world’s leading search scientists and a pioneer in the field of collaborative exploratory search.

Research Sponsorship

In addition to serving on CISML's Industrial Advisory Board, Catalyst is helping to sponsor its research. The relationship allows Catalyst to collaborate with CISML faculty and students on research that is of mutual interest. By virtue of the relationship, Catalyst has access to public domain software developed within the center and exposure to top graduate and undergraduate students for possible internships or hires.