Back to News

H5 Granted Patent for Enhanced Method of Clustering Datasets

H5 Granted Patent for Enhanced Method of Clustering Datasets

SAN FRANCISCO, CA – October 16, 2013

H5, a leading provider of technology-assisted review and e-discovery services, has patented its unique system and method for clustering large datasets. Granted by the U.S. Patent & Trademark Office, patent number US 8,515,956 covers a method for more efficiently identifying and organizing data in large datasets that optimizes system resources, enabling faster and more scalable clustering on large datasetsthan more conventional clustering approaches.

The volume of data implicated in litigation and investigations today is growing rapidly. Clustering is a general method for grouping similar documents together, making it easier to manage large datasets and identify important relationships between documents. Conventional clustering systems that convert data elements into numerical values require significant time and processing resourcesto assess the similarity of data elements. H5’s patented method utilizes a novel clustering methodology that insteadtakes advantage of the efficiency of bit operations on bit vectors, reducing the time and resources required to establish data relationships. H5 uses the patented method for grouping like documents together in data management scenarios, such as document productions for litigation or data mining, when datasets might include text documents, audio files, video files or image files. Reducing the time it takes to cluster documents enhances the categorization process for large data volumes that H5 handles in its rules-based technology-assisted review for litigation, enabling quick turnaround for counsel and lower costs for the client while ensuring consistency and reliability.

“With so much electronic information to process in litigation, quick turnaround can often make or break a case,” said Jeff Kangas, Executive Managing Director.“This latest patent is a testament to our team’s continued innovation and our commitment to the development of tools and methods that can address the deadline-driven services we provide to our clients while maintaining the high precision and high recall accuracy that is our stock in trade.” The unique clustering method augments H5’s already patented high precision, high recall document review process and adds to H5’s growing portfolio of patents that cover a range of systems and methods for organizing, assessing and searching the growing volume of information being generated by organizations today. H5 primarily utilizes the patented technologies and methods to help corporate counsel and law firms address document review and production obligations in document intensive litigation, investigations and compliance. The patented methods have a broad range of potential uses in any field that deals with large volumes of documents or data.

About H5

H5, a leading provider of technology-assisted review and e-discovery services, is dedicated to finding information that clients need to win cases, meet regulatory requirements and address risks before they become liabilities. By combining advanced technologies with expertise in search and information analysis, H5 delivers customized solutions to help corporations and law firms meet their legal discovery, compliance and information management objectives. Numerous independent studies have proven H5’s approach delivers greater cost savings and more accurate results than any alternative. See for more information.


Thank you for subscribing to the H5 blog, True North.

We strive to provide quality content on a variety of topics related to search, eDiscovery and the legal realm.

Please check your email inbox for your subscription confirmation!