Artificial Intelligence in Law (and beyond) including Machine Learning as a Service, Quantitative Legal Prediction / Legal Analytics, Experts + Crowds + Algorithms
1.
a six part primer
artificial intelligence in law (and beyond)
daniel martin katz
blog | ComputationalLegalStudies.com
corp | LexPredict.com
page | DanielMartinKatz.com
edu | chicago kent college of law
lab | TheLawLab.com
[ a.i. + law ]
2.
3. There has been lots of recent
interest in applying
artificial intelligence to law
4.
5.
6.
7.
8.
9.
10.
11.
12.
13. and there is a bit of confusion
as to where we stand today
and where we are headed
17. data driven AI rules based AI
Competing Orientations in
Artificial Intelligence
18. expert
systems
Computational Law
Data Driven Rules Based
prediction
models
and
methods
network
analytic
methods
natural
language
processing
self
executing
law
visual
law
computable
codes
19.
20. we see a decent amount of
rules based AI
in legal industry
32. JUSTICE GAP
80%Civil legal needs of
low-income people in
the U.S. go unmet
For every 1 person
served in an LSC-
funded program, at
least 1 person is
turned away
33. LSC TECH
SUMMIT
“to explore the potential of
technology to move the United States
toward providing some form of
effective assistance to 100% of
persons otherwise unable to afford an
attorney for dealing with essential
civil legal needs.”
34. LSC TECH
SUMMIT
“to explore the potential of
technology to move the United States
toward providing some form of
effective assistance to 100% of
persons otherwise unable to afford an
attorney for dealing with essential
civil legal needs.”
some form of effective
assistance to 100% !
41. JUSTICE &
TECHNOLOGY
PRACTICUM
STUDENT WORK
!
Fieldwork (e.g. Self-Help
Web Center)
Scope document
Research memo
Storyboard
A2J Guided Interview & HotDocs
template
Final presentation
Professor
Ron Staudt
IIT Chicago-
Kent College
of Law
Engage community partners: legal aid
organizations, courts
47. Client Intake
Client Acquisition
More Seamless Client Interaction
via Tech Platform
Providing Legal Information
to Non-Lawyers in Large
Organizations
48.
49. so although we see a
decent amount of
rules based AI
in legal industry
50. I am pretty bearish
on Rules Based A.I.
for most applications …
51. my views are informed by
the history of A.I. in general
52. lots of issues
with expert systems
and/or
rules based A.I.
(without data or an evolutionary dynamic)
53. rules based A.I. data driven A.I.
1980’s, 1990’s, Early 2000’s
>
54. rules based A.I. data driven A.I.
1980’s, 1990’s, Early 2000’s
>
rules based A.I. data driven A.I.
2005 - Present
<
~
55. Ultimately we are trying to learn
the rules / dynamics that
underlie some class of activity
59. expert
systems
Computational Law
Data Driven Rules Based
prediction
models
and
methods
network
analytic
methods
natural
language
processing
self
executing
law
visual
law
computable
codes
113. General Counsels as Legal
Procurement Specialists
TyMetrix/ELM -
Using $50 billion+ in Legal
Spend Data to Help GC’s
Look for Arbitrage
Opportunities, Value
Propositions in Hiring Law
Firms
#LegalSpendAnalytics
Quantitative Legal Prediction
129. Columbia Law Review
October, 2004
Theodore W. Ruger, Pauline T. Kim,
Andrew D. Martin, Kevin M. Quinn
Legal and Political Science
Approaches to Predicting
Supreme Court Decision
Making
The Supreme Court
Forecasting Project:
161. “Software developers were asked on
two separate days to estimate the
completion time for a given task, the
hours they projected differed by 71%,
on average.
W h e n p a t h o l o g i s t s m a d e t wo
assessments of the severity of biopsy
results, the correlation between their
ratings was only .61 (out of a perfect
1.0), indicating that they made
inconsistent diagnoses quite frequently.
Judgments made by different people
are even more likely to diverge.”
167. (most pundits did not
identify as a serious
candidate him until
mid-January 2017)
Neil Gorsuch was #1
o n o u r F a n t a s y
Platform 12 Days after
Donald Trump was
elected President
(i.e Nov 20)
171. we have developed an
algorithm that we call
{Marshall}+
random forest
172. Columbia Law Review
October, 2004
Theodore W. Ruger, Pauline T. Kim,
Andrew D. Martin, Kevin M. Quinn
Legal and Political Science
Approaches to Predicting
Supreme Court Decision
Making
The Supreme Court
Forecasting Project:
173. Ruger, et al (2004)
relied upon
Brieman(1984)
(as partially shown below)
179. Random forest is an approach to
aggregate weak learners into
collective strong learners
(using a combo of bagging and random substrates)
(think of it as crowd sourcing of models)
184. Final Version of #PredictSCOTUS
1816-2015
case accuracy
70.2%
71.9%
justice accuracy
http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0174698
195. expert
forecast
crowd
forecast
learning problem is to discover how to blend streams of intelligence
algorithm
forecast
ensemble method
ensemble model
via back testing we can learn the weights
to apply for particular problems
196. By the way, you
might ask why does
one care about
marginal improvements
in prediction ?
#Fin(Legal)Tech
197. Given our ability to offer
forecasts of judicial
outcomes, we wondered
if this information could
inform an event driven
trading strategy ?
198. Revise + Resubmit @
http://arxiv.org/abs/1508.05751
available at
http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2649726
207. Fin(Legal)Tech Conference
finlegaltechconference.com
@Illinois Tech - Chicago Kent College of Law
Fin(Legal)Tech Conference
finlegaltechconference.com
@Illinois Tech - Chicago Kent College of Law
Fin(Legal)Tech Conference
finlegaltechconference.com
@Illinois Tech - Chicago Kent College of Law
290. Daniel Martin Katz
@ computational
computationallegalstudies.com
lexpredict.com
danielmartinkatz.com
illinois tech - chicago kent college of law@
thelawlab.com