Delivers More Than 99 Percent Foreign Character Recognition Accuracy
SAN FRANCISCO, CA -- (MARKET WIRE) -- 07/23/07 -- Discovery Mining,
a leading provider of Automated Discovery software, today announced the availability of the industry's first encoding-agnostic foreign language functionality, enabling customers to process high volumes ofdata in multiple languages and character sets with more than 99 percent recognition accuracy.

In today's environment, organizations are dealing with text and symbols of numerous languages from around the world, which need to be represented and manipulated in a consistent manner for accurate data retrieval during the electronic discovery process. Although Unicode standards have been implemented to help relieve these issues, users of traditional electronic discovery Unicode functionality often find there are variations in how data encoding is handled. These variations can lead to inaccurate data; for example, an email in
Japanese may contain an attachment created with non-Unicode standards, as well as another attachment in Korean, encoded in a third method. To ensure that encoding inconsistencies do not affect the end-data, companies often enlist native language speakers to provide quality control of all documents, making sure they are both consistent and readable. Naturally, this can be an expensive and time-consuming process. Utilizing encoding-agnostic foreign languagefunctionality, when files are larger than 200 bytes, Discovery Mining
achieves greater than 99 percent accuracy, without native speaker
supervision.

"Though great strides have been made in the foreign character standards, users continue to find a lack of integrity and accuracy in their data sets during the electronic discovery process," said Matthew Work, CEO of Discovery Mining. "Discovery Mining addresses this pain point with software that elegantly handles multiple languages as part of a single, efficient process -- a benefit no other vendor offers."

The unique algorithms contained in Discovery Mining's Unicode functionality solve the problem of inconsistency by analyzing each file, automatically identifying the encoding and languages used, converting it to the correct encoding format and then processing it with Discovery Mining's Unicode capabilities. By automating the process, Discovery Mining removes the need for human intervention, reducing both cost and complexity.

About Discovery Mining, Inc.
Discovery Mining is a leading provider of Automated Discovery
software that stores, categorizes, searches and presents documents to
organizations involved in litigation, investigation, regulatory
response, or mergers & acquisitions. The Company's proprietary
technology significantly reduces risk and cost at every point in the
Document Lifecycle. Trusted by top law firms including Am Law 100,
and by Fortune 1000 corporations, Discovery Mining helps
organizations achieve and maintain excellence in Web-based automated
search and retrieval document management. Headquartered in San
Francisco, Discovery Mining also has offices in New York City and
London, and is present in Chicago and Washington, D.C. For more
information call 1-866-343-4205, email info@discoverymining.com, or
visit www.discoverymining.com. To view Discovery Mining co-founder
Andrew Jenks' blog, visit www.discoverymining.com/dmblog/.