This Page

has been moved to new address

Do Litigators Need to Understand Predictive Coding Theory?

Sorry for inconvenience...

Redirection provided by Blogger to WordPress Migration Service
----------------------------------------------------- Blogger Template Style Name: Snapshot: Madder Designer: Dave Shea URL: / Date: 27 Feb 2004 ------------------------------------------------------ */ /* -- basic html elements -- */ body {padding: 0; margin: 0; font: 75% Helvetica, Arial, sans-serif; color: #474B4E; background: #fff; text-align: center;} a {color: #DD6599; font-weight: bold; text-decoration: none;} a:visited {color: #D6A0B6;} a:hover {text-decoration: underline; color: #FD0570;} h1 {margin: 0; color: #7B8186; font-size: 1.5em; text-transform: lowercase;} h1 a {color: #7B8186;} h2, #comments h4 {font-size: 1em; margin: 2em 0 0 0; color: #7B8186; background: transparent url( bottom right no-repeat; padding-bottom: 2px;} @media all { h3 { font-size: 1em; margin: 2em 0 0 0; background: transparent url( bottom right no-repeat; padding-bottom: 2px; } } @media handheld { h3 { background:none; } } h4, h5 {font-size: 0.9em; text-transform: lowercase; letter-spacing: 2px;} h5 {color: #7B8186;} h6 {font-size: 0.8em; text-transform: uppercase; letter-spacing: 2px;} p {margin: 0 0 1em 0;} img, form {border: 0; margin: 0;} /* -- layout -- */ @media all { #content { width: 700px; margin: 0 auto; text-align: left; background: #fff url( 0 0 repeat-y;} } #header { background: #D8DADC url( 0 0 repeat-y; } #header div { background: transparent url( bottom left no-repeat; } #main { line-height: 1.4; float: left; padding: 10px 12px; border-top: solid 1px #fff; width: 428px; /* Tantek hack - */ voice-family: "\"}\""; voice-family: inherit; width: 404px; } } @media handheld { #content { width: 90%; } #header { background: #D8DADC; } #header div { background: none; } #main { float: none; width: 100%; } } /* IE5 hack */ #main {} @media all { #sidebar { margin-left: 428px; border-top: solid 1px #fff; padding: 4px 0 0 7px; background: #fff url( 1px 0 no-repeat; } #footer { clear: both; background: #E9EAEB url( bottom left no-repeat; border-top: solid 1px #fff; } } @media handheld { #sidebar { margin: 0 0 0 0; background: #fff; } #footer { background: #E9EAEB; } } /* -- header style -- */ #header h1 {padding: 12px 0 92px 4px; width: 557px; line-height: 1;} /* -- content area style -- */ #main {line-height: 1.4;} {font-size: 1.2em; margin-bottom: 0;} a {color: #C4663B;} .post {clear: both; margin-bottom: 4em;} .post-footer em {color: #B4BABE; font-style: normal; float: left;} .post-footer .comment-link {float: right;} #main img {border: solid 1px #E3E4E4; padding: 2px; background: #fff;} .deleted-comment {font-style:italic;color:gray;} /* -- sidebar style -- */ @media all { #sidebar #description { border: solid 1px #F3B89D; padding: 10px 17px; color: #C4663B; background: #FFD1BC url(; font-size: 1.2em; font-weight: bold; line-height: 0.9; margin: 0 0 0 -6px; } } @media handheld { #sidebar #description { background: #FFD1BC; } } #sidebar h2 {font-size: 1.3em; margin: 1.3em 0 0.5em 0;} #sidebar dl {margin: 0 0 10px 0;} #sidebar ul {list-style: none; margin: 0; padding: 0;} #sidebar li {padding-bottom: 5px; line-height: 0.9;} #profile-container {color: #7B8186;} #profile-container img {border: solid 1px #7C78B5; padding: 4px 4px 8px 4px; margin: 0 10px 1em 0; float: left;} .archive-list {margin-bottom: 2em;} #powered-by {margin: 10px auto 20px auto;} /* -- sidebar style -- */ #footer p {margin: 0; padding: 12px 8px; font-size: 0.9em;} #footer hr {display: none;} /* Feeds ----------------------------------------------- */ #blogfeeds { } #postfeeds { }

Thursday, May 10, 2012

Do Litigators Need to Understand Predictive Coding Theory?

With the debate over Predictive Coding entering a feverish pitch, there is an interesting thread of discussion beginning to emerge asking whether or not litigators and other users need to understand what I am going to refer to as Predictive Coding Theory.

In a May 4, 2012 Blog  titled, "3 Drawbacks To Predictive Coding", Sandra E. Serkes, President and CEO of Valora Technologies writes "What is missing (in regards to the Predictive Coding debate), is a discussion of the specific weaknesses of the overall Predictive Coding technique.   She then goes on to indicate that, "Predictive Coding tagging algorithms are not transparent".

To put this into more technical terms, do we need to know what probability theories and related  dimension reduction systems are being used as the foundational algorithms for Predictive Coding system to identify relevant documents?

For those of you who are interested in a more detailed overview of Predictive Coding, I suggest that you read a March 25, 2012 Blog post titled, "Predictive Coding Based Legal Methods for Search and Review", Ralph Losey does an excellent job of discussing the basic technical mechanics and some of the underlying theories of Predictive Coding.

Getting back to my question about how much we need to know about Predictive Coding, I am in the process of developing some unique insight.  Over the past 30 days, in preparation for adding a Predictive Coding module to the DCIG/eDSG 2012 Early Case Assessment Buyers Guide, I have been interviewing product managers from some of the Predictive Coding vendors and current users of Predictive Coding system to develop a list of criteria for reviewing and ranking the platforms for our buyers guide.  One of the questions that I have been asking is what probability theories and related  dimension reduction systems are being used as the foundational algorithms for your Predictive Coding platform to identify relevant documents.

So far, I haven't gotten a straight answer as most of the product managers either don't undesrtand the question or want to move the discussion up a couple of layers in technology stack to talk about indexing, semantic search, clustering, relevance ranking, sampling and presentation of results.  There is no doubt that these are all very pertinent topics to a perspective buyer of Predictive Coding technology.  However, it doesn't answer the question about the transparency of exactly how these systems are identifying relevant documents.

Whether or not litigators need to understand Predictive Coding theory and the underlying probability theories and related  dimension reduction systems is debatable.  However, I believe that a minimum level of transparency from the Predictive Coding vendors would at least give buyers the opportunity to understand what they are buying and then compare the various offerings.

In 1969, Edgar F. Codd and some of his associates that I have actually had the honor or knowing,  first formulated and proposed the theory of relational database.  And, although I am not sure that it reach the level of skepticism and resistance to blind adoption that we are currently seeing with Predictive Coding.  However, it was new and therefore many did require an explanation of the underlying theories and mathematics.  Eventually,  the discipline normalized and everyone just assumed that relational databases worked and there was no longer any need to question "how they worked'.

A similar vetting process would be very healthy for Predictive Coding.  Check back to my blog in the coming weeks for updates and more information on this topic.

Labels: , , , , , ,


At May 23, 2012 at 3:42 PM , Blogger Unanymous said...

We have always received new technology with a pinch of salt. Every technology has its own period to stabilize before it can be put to good use. Secondly expecting transparency from software vendors is equivalent to asking Google to publish the details of their search algorithm(which is never going to happen!!!!). Its up to experts and researchers to test the various tools and provide unbiased result comparison from these tools that provide predictive coding review support.

At May 23, 2012 at 10:40 PM , Blogger phatik said...

My cousin recommended this blog and she was totally right keep up the fantastic work!

Litigation Coding


Post a Comment

Subscribe to Post Comments [Atom]

<< Home