In the great work of Alexandre Dumas, "The Count of Monte Cristo," our protagonist spends 30 years trapped in an island dungeon, makes a daring escape and, using the knowledge he gathered during his time in prison, finds treasures beyond imagination awaiting him just where he was told they would be. (He then goes off to seek revenge on those who did him wrong, but let's skip that part.)
Much of the legal profession seems to be similarly trapped—or at least ensnared in place —for the past several decades, still content to stick to tried and true methods to search for larger and larger amounts of electronically stored information, relying on linear, manual review, as well as keyword search techniques.
However, I will assume that readers of this column are familiar with U.S. District Court Magistrate Judge Andrew Peck's landmark decision in Da Silva Moore v. Publicis Groupe SA et al , 287 F.R.D. 182 (S.D.N.Y. 2012) announcing that "predictive coding" and similar technology-assisted review methods should be considered as a reasonable alternative to traditional keyword searching in appropriate cases.
Indeed, one cannot go very far these days without encountering an e-discovery CLE program that has a TAR session or track. I am serene in my belief that it is only a matter of time before a much greater proportion of the legal community sees the value in using TAR methods in more and more cases.
This is good, but doesn't in my book go far enough.
The proposition I'd like to advance here is that lawyers should be aware of the real (and greater) secret revealed in all the recent research and scholarship leading up to Da Silva Moore , namely, that TAR-like methods have multiple uses beyond simply determining relevance or privilege for large data sets.
That is, if we have been paying close attention to the results of studies coming out of the Text Retrieval Conference ( TREC ) Legal Track and the Electronic Discovery Institute, we should have learned from our colleagues in computer and information science (who are familiar with analytics employing text mining and machine learning) that powerful techniques exist to better exploit data sets.
These techniques, which include TAR-like methods, in turn hold the potential to pay off in handsome dividends to our legal practice.
I am talking, of course, about analytical methods used in ways to optimize information governance.
Thomas Davenport and Jinho Kim, in their 2013 book "Keeping Up with the Quants : Your Guide to Understanding and Using Analytics," define "analytics" to mean "the extensive use of data, statistical and quantitative analysis, explanatory and predictive models, and fact-based management to drive decisions and add value."
No comments:
Post a Comment