INSIGHT
Knowledge: people & systems
‘‘
ll systems are dynamic, and knowledge systems especially so, not
only in the narrow sense – the conjunction of the code and the machine – but also in a wider sense: the people that support and use a system, and the processes this involves. Over the course of my career at Linklaters, my role has often shifted following changes in our knowledge systems, but I have also been able to change the systems themselves. After completing a Master of Library and Information Science degree in Canada, followed by time spent at the University of Tokyo, I joined Linklaters’ Tokyo offi ce as an Information Offi cer in 2006. My role was divided between helping foreign and Japanese lawyers with research, and indexing memoranda and other documents as legal knowledge. Both tasks were connected to a system called Dealfi nder, which aimed to serve as the fi rm’s central repository of knowledge. Dealfi nder and its predecessors were ahead of their time but shared a number of limitations. For example:
● They were standalone systems, so our information teams needed to re- type tremendous amounts to ensure that we had minimal coverage of the work done by the fi rm.
● We provided links to the fi nalised matter documentation but the documents we added were not searchable, meaning expensive time wasted by lawyers.
● Each record ‘belonged’ to one of the fi rm’s internal Practice Groups, which led to the creation of duplicate records as teams tried to capture the cross- practice nature of their work.
● Each record could have only one, English-language example of credential text, at odds with the needs of our regional marketing teams to
March 2019
Experiments with artifi cial intelligence have proven benefi cial for creation of new search fi lters, but they have also demonstrated there are no magic wands.
A look under the bonnet A
deliver pitches to prospective clients around the world, in the language of their choice.
Many of the frustrations of my early role as an Information Advisor in Tokyo were based on the shortcomings of these systems, and my fi rst attempts at learning to code focused on writing (quite primitive!) Excel add-ins that helped me – and others – work around them.
Moving on
Fast forward a few years, and after completing a part-time Masters in Management of Information Systems, I found myself a Knowledge Systems Manager based in Linklaters’ London offi ce, eager to improve upon the very systems which had caused me much frustration years earlier. The technological landscape has changed and our new system, MatterExplorer, is built on a platform that makes use of linked data, graph databases, and artifi cial intelligence. We can now cross-reference fi nancials with the richer legal metadata stored in the legacy systems, supporting pricing discussions and opening the possibility for the development of more targeted pricing tools in the future. It also means that our information teams no longer need to spend valuable time simply re-entering data.
Powerful search
Linked data has also enhanced our document search. The usefulness of simple full-text search for legal texts is often being undermined by the repetitive, highly-structured nature of the documents. With MatterExplorer we can build up fi lters based on the records to which the documents are connected. This gives Linklaters’ information professionals powerful tools for supporting our lawyers in running searches. Experiments with artifi cial intelligence have proven benefi cial
James Pilgrim (
james.pilgrim@linklaters.com) is Senior Knowledge Systems Manager, Linklaters.
for the creation of search fi lters, but they have also demonstrated there are no magic wands. There is considerable eff ort and expertise required to defi ne documents types of interest, break these into component clauses, and train and test the eff ectiveness of any AI tool. Here, again, information professionals have a role as an intermediary between technical and legal teams.
Key role
MatterExplorer has already made a diff erence in the day-to-day work of many people at Linklaters, but as we approach the launch of its second phase there is much to do. The technologies mentioned above have addressed a number of limitations, but each of these, once overcome, has led to questions, opportunities to develop the system in ways and to achieve the previously unachievable.
At each step, information professionals have a key role to play in defi ning what needs to be done and how we can do it and I will explore some of this in my next column. IP
INFORMATION PROFESSIONAL 19
Data Systems
p19.indd 1
07/03/2019 10:52
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48 |
Page 49 |
Page 50 |
Page 51 |
Page 52 |
Page 53 |
Page 54 |
Page 55 |
Page 56 |
Page 57 |
Page 58 |
Page 59 |
Page 60
