1 00118601 Emerging themes 2019 A4 AW v31 combined - Page 51

Continuous active learning
Another form of predictive coding allows you to use the
traits within certain relevant or “hot” documents to apply
a relevancy score to all the documents. The reviewer can
then choose to review only those documents which have
a relevancy score above a certain limit (although they
should also review samples of documents scored below
this limit as a control). This approach is commonly known
as “continuous active learning”.
Sentiment analysis and other non-content analysis
We are now seeing tools being created which analyse
“sentiment” or the tone in which something has been
communicated. In addition to this “sentiment analysis”
there have been advancements in analysing and
searching non-content files such as video, audio and
images. These tools are intended to identify specific
objects, such as the presence of a mobile phone on a
trading floor. This will ensure that these files are not just
“swept-aside” during the review, and the review of such
files are focused rather than having to review every
frame or second of audio.
and process. It represents a rethink of how to conduct
a preliminary investigation or early case assessment,
using a range of forensic tools to appraise up to 5GB of
data far more quickly and effectively than a traditional
review allows. This, in turn, allows us to very quickly
prepare recommendations, with a greater level of
insight than would otherwise be possible to achieve
at proportionate cost.
In summary
The days of increasing costs and human time spent
reviewing and analysing data are well and truly over.
The age of the machine is here, and the tools that the
technology is giving rise to continue to make large scale
document reviews increasingly efficient and effective.
BCLP Clear/Cut
Having access to cutting-edge tools is not enough;
it requires a skilled team of technologists and legal
experts to work closely together in training the machine
learning algorithms effectively. BCLP’s recently launched
Clear/Cut service is a prime example of what can be
achieved by maximising the roles of technology, people
Associate, London
Forensic Technology
Manager, London


Powered by

Full screen Click to read
Paperturn flip book
Download as PDF
Shopping cart
Full screen
Exit full screen