ZyLAB uses text analytics in eDiscovery

April 14, 2010 — 9:01am ET | By Ron Miller

Tools

Tags
ZyLab
eDiscovery
Text Analytics

Text analytics appear to be going mainstream. This week, ZyLAB announced it was incorporating text analytics capabilities into its eDiscovery product, just a week after Open Text announced it was adding text analytics from its NStein purchase across its ECM properties. Up until fairly recently text analytics was a fairly esoteric, even academic undertaking, but now we are seeing it applied inside the enterprise to help companies better understand high volumes of information and find patterns that might not otherwise be obvious with conventional analysis methods.

The hope is that by applying text analytics as part of the eDiscovery process, it will lower the overall cost associated with finding and classifying information. According to ZyLAB:

"Text analytics enable organizations to identify, classify, and sort relevant enterprise information prior to initiating the formal Electronic Discovery Reference Model (EDRM) Collection process and extract entities, facts, relationships, concepts and subtext that will influence the litigation strategy. Legal counsel gains greater insight into potential liabilities and can make informed decisions much earlier in the process."

The tool helps with among other things, auto-classifying, extraction of 200 data type entities, fact extraction and analysis of unknown patterns or trends. With any text analytics products, you need a leap of faith that the system has been well tuned to get at the information you need without generating lists of information types that aren't useful to your team.

If the text analytics capability can help bring order to the information analysis process and lower the overall cost of eDiscovery by speeding up the process, it will be a big boon to those charged with pulling this information together in a legal or regulatory action.

For more information:
- see the ZyLAB press release