An Electronic Discovery Blog covering News, Articles
and Thoughts for the Legal and Corporate Community Author: Alexander H. Lubarsky, LL.M., Esq. - email@example.com - Tel. (415) 533-4166 OR 800-375-4222 THIS BLAWG IS NOT AFFILIATED WITH THE WEB SITES WWW.DISCOVERYRESOURCES.ORG OR WWW.DISCOVERYRESOURCES.COM
Friday, January 21, 2005
Attached to Attenex Here is a very thoughtful and well supported response to my previous call for info regarding people's thoughts on concept based querying. This response comes from Jim Sherman of the firm Preston, Gates and Ellis in Seattle and truly lends a clear understanding of what concept based technologies are and where they are likely to head...
Here you go... my beloved Blawgerheads, enjoy....
Alex, Hi. I’m a partner with the law firm of Preston Gates & Ellis LLP, and a member of Preston’s Document Analysis Technology Group (DATG). I’m happy to be able to comment on the issues and questions that you raise below.
I should start by disclosing, as I think you are aware, that Attenex Corporation—the company that developed and markets Attenex Patterns E-Discovery Platform—was formed and is owned by Preston.
Preston has worked extensively on document discovery matters over the past 10+ years, and has experienced first hand the shift of document discovery from almost pure paper to today’s focus on K:\19406\00045\JDS\JDS_O210Je-discovery. Several years ago, as we were struggling to find an effective way to efficiently and economically organize and review the mushrooming volume of electronic documents, we hit upon the idea of using concept based searching/organization. At that time, there were no concept-based tools available that had been designed specifically for use by the legal community, much less specifically for document discovery. Because there was no such technology available, the firm formed Attenex Corporation to develop the software we needed.
Given the above information, it will be no surprise to you that my direct, hands-on experience with using concept-based technology is pretty much confined to the Attenex e-discovery platform.
Before turning to your specific questions, I think it’s important to note that concept-based document discovery technology (at least insofar as the Patterns tool is concerned) is not a stand-alone product. Rather, it is one important component of a tool that employs various other technologies for indexing, searching, de-duplicating, and graphic display. In our experience, the most valuable aspect of Patterns’ concept-based technology is not its ability to retrieve documents using concept-based searching, but rather how the software organizes documents in a graphic display based on the concepts they contain. It is really the concept based organization and display, rather than the concept-based searching, to which my comments below refer.
We have used Attenex’s e-discovery software for over 3 years. The product integrates keyword searching and de-duplication with concept-based technology together in an interface that allows the user to re-organize, view, and mark documents. The software includes Boolean search capability, which we use quite extensively, and which is an important complement to the concept-based organization of documents provided by the tool.
We use Boolean and concept searching to identify the documents on which we want to focus, and then use the concept based technology to organize the documents based on their actual conceptual content. So, although the initial searches may pick up “red herrings,” those documents will tend to be grouped together and located in one area of the graphic display, while the relevant documents (which likely have different conceptual content) will be displayed in a different area. This organization allows us to more quickly distinguish between relevant and irrelevant documents – so that in addition to more efficiently locating and marking responsive documents, we can zero in on key or important documents, while not wasting time on irrelevant materials.
We are big fans of the technology. Our DATG uses the technology in virtually all of its cases involving electronic discovery. We feel it makes a big difference to our group. It has substantially increased our productivity, as well as the quality and consistency of our reviews, while saving our clients money.
Finally, I do believe that the use of concept-based technology is here to stay. Traditional searching alone, including Boolean searching, is inadequate to effectively plow through the massive amounts of electronic data we are seeing in e-discovery.
Hope this is of some use to you. Let me know if you have any questions or comments.
posted by Alexander | 5:38 PM