As a result, they turned to Technology Assisted Review (TAR) and
Predictive Coding systems to locate relevant data.
And that's before we start to venture into the territory of the top end of eDiscovery software and the vendors supplying
predictive coding systems based on NLP and machine learning.
Most
predictive coding systems require human reviewers to train and tune them.
Not exact matches
For instance, in «Document Review», a lot of time is to be saved in reviewing by using
predictive coding, but some of the gains must be reinvested in classifying an appropriate sample and in training the
system.
As
predictive coding has evolved into «technology - assisted review,» machine learning and other AI tools have created some amazing results that offer ways to improve that whole area of practice, save costs and help clients and the court
system.
Predictive Analytics (also called «
Predictive Coding» or «Technology Assisted Review») is a workflow that requires a subject matter expert to review a small subset of documents in order to train the
system on what the human is looking for until the
system can statistically «predict» how the human would
code the rest of the collection.
Finally, and perhaps not obviously at all, lawyers also work with technology — not simply word processing and email, but software for document generation, electronic discovery,
predictive coding and technology assisted review, analysis by rules - based expert
systems, blockchain.
Using our «Expert
Systems» for automated issue
coding (also called
predictive coding) makes a huge difference to plaintiffs» counsel in terms of cost, quality and time spent for high - volume document review projects.
In the end, TAR review may end up costing more than manual review due to poorly implemented
predictive coding protocols, review results disputes between the parties, or the desire to put large volumes of data through the
system simply because it was designed to handle it.
Unlike in eDiscovery, where
predictive coding won't work out of the box, our
system comes with lots of knowledge built in.
A lot of
predictive coding tools employ a relevancy score, which essentially rates the
system's confidence on a document's relevance.
Machine learning and
predictive coding in e-discovery are only the beginning of technology's impact on the legal
system: Law is the ultimate digital product, and this will be borne out in time.
• Ipro Tech reports that it is adding
predictive coding, concept searching, and other features to its Eclipse discovery
system.
He cited his
predictive coding experience: several years ago, it took him 15,000 to 20,000 records to train a
system.