addressalign-toparrow-leftarrow-rightbackbellblockcalendarcameraccwcheckchevron-downchevron-leftchevron-rightchevron-small-downchevron-small-leftchevron-small-rightchevron-small-upchevron-upcircle-with-checkcircle-with-crosscircle-with-pluscrossdots-three-verticaleditemptyheartexporteye-with-lineeyefacebookfolderfullheartglobegmailgooglegroupshelp-with-circleimageimagesinstagramlinklocation-pinm-swarmSearchmailmessagesminusmoremuplabelShape 3 + Rectangle 1ShapeoutlookpersonJoin Group on CardStartprice-ribbonShapeShapeShapeShapeImported LayersImported LayersImported Layersshieldstartickettrashtriangle-downtriangle-uptwitteruserwarningyahoo

Predicting and Understanding Law with Machine Learning

For our August Data Science DC Meetup, we are excited to have John J. Nay,  a Ph.D. Candidate at the Vanderbilt University School of Engineering and a Research Fellow at Vanderbilt Law School’s Program on Law & Innovation, join us to speak about predicting and understanding law with machine learning!

---------------------------- 

Sponsored by Statistics.com -- Learn new skills in Data Science, Analytics, and Statistics! Four-week online courses at all levels, world-class instructors, private discussion boards. Save 30% with code DSDC2016.

----------------------------

Agenda:

• 6:30pm -- Networking, Empanadas, and Refreshments

• 7:00pm -- Introduction, Announcements

• 7:15pm -- Presentation and Discussion

• 8:30pm -- Data Drinks (Tonic , 2036 G St NW)

Abstract:

Predicting and Understanding Law with Machine Learning

First, we present new methods to embed institutions and their legal text into shared continuous vector space to enable novel investigations into differences across institutions. Our model discerns meaningful differences between government branches. The similarities between learned representations of Congresses over time and sitting Presidents are negatively correlated with the bill veto rate, and the temporal ordering of Presidents and Congresses was implicitly learned from only text.

Second, we describe PredictGov, a machine learning system for predicting Congress. Starting with the 107th Congress, we trained models on data from previous Congresses, predicted all bills in the current Congress, and repeated until the 113th Congress served as the test. The median of the predicted probabilities for enacted bills was 0.71, while the median of the predicted probabilities for failed bills was 0.01. We also conducted a sensitivity analysis on the model to determine important factors predicting enactment. We will describe these results and demonstrate its interactive implementation for real-time predictions of the 114th Congress.

Bio:

John J. Nay

John J. Nay is a Ph.D. Candidate at the Vanderbilt University School of Engineering and a Research Fellow at Vanderbilt Law School’s Program on Law & Innovation. He received his bachelor’s degree with High Distinction from the University of Virginia. His computational work is directed toward law and policy applications, including computer simulations of climate prediction markets, software for forecasting drought globally with satellite data, computational models predicting human cooperation, software for automatically estimating models of decision-making, natural language processing of law and policy, and machine learning for predicting and understanding law-making. He is currently collaborating with social scientists, environmental scientists, and engineers on multiple National Science Foundation-funded projects, and with lawyers on developing legal technology. More information can be found at johnjnay.com.

Sponsors:

This event is sponsored by the George Washington Business School MS in Business Analytics ProgramStatistics.comElder Research, NovettaPAWGOV, O'ReillyBooz Allen Hamilton, and AOL. (Would your organization like to sponsor too? Please get in touch!) 

Join or login to comment.

  • JNay

    Thank you everyone for coming out! I've posted the slides from my presentation here: http://johnjnay.com/presentations/

    5 · August 3

    • Stephen H.

      thanks, John! it is one of the best presentations i listened to recently

      1 · August 3

  • Stephen H.

    Great Talk! A nice combination of NLP, machine learning, law, and politics!

    2 · August 1

    • Stephen H.

      btw, will the slides be uploaded?

      August 2

    • Doug_S

      Maybe we can compute a probability of that? ;-)

      1 · August 2

  • carla

    I enjoyed the talk and would like to learn more about ML. Am looking for a tutor who can teach me ML, networking and cluster analysis for a public policy research project...

    1 · August 2

    • carla

      Thanks for the info!! I can pay $50/hr for in-person tutoring by the GWU or Farragut area. Thanks.

      1 · August 2

    • Sean Moore G.

      Great Carla, please email us at [masked]

      August 2

  • Nevin H.

    Tough ML chore, but excellent presentation of all efforts, well done

    1 · August 2

  • Ann V.

    Thx to John Jay for great ML use case prezi - ck out johnjnay.com for follow-up deep dive info generously shared - also big shout-out to Ian Balin/IBM for sponsoring includ swell follow-up at Tonic! :-) - and last but not least ongoing thx to Organizers for making it all happen and keeping this group going so super!!!

    3 · August 2

  • Doug_S

    As is, we were left somewhat unsure of what Dr. Nay's approach is telling us about what controllable factors affect a bill's chances, and not a bit wiser about how to tell what the proposed legislation would actually do. As the old saying goes, further research is indicated. Meanwhile, to paraphrase another old saying (by Aaron Levenstein, about statistics), what we got is like a bikini: what is revealed is suggestive, but what is concealed is vital.

    August 2

  • Doug_S

    Very good presentation, somewhat curious focus. I wouldn't have guessed that the chance of enactment of proposed legislation would be the main topic. As Dr. Nay mentioned, enacted bills often get lengthy, with possibly conflicting provisions -- the Affordable Care Act, at nearly 2000 pages, is a prime example. There should be a splendid opportunity for machine learning and natural language processing in helping people understand what a law actually says, in light of precedental decisions. Changing understanding of what language provides might, in turn, influence its chance of enactment -- as well as its chance of surviving judicial review.

    1 · August 2

  • Chris G.

    Very interesting! Thank you for sharing your insights!

    August 1

  • Ian

    We're outside on the patio behind

    August 1

  • Sean Moore G.

    Predictgov.com

    4 · August 1

  • Tom L.

    Predictiveanalyticsworld.com is offering a vendor-neutral government consulting analytics conference. For more information visit their link above. Use code DCDC for 10% off

    August 1

  • Tom L.

    For Access to online data and information reports as well as information on seminars in the field of data science please visit our sponsor's site at O'Reilly.com/go/ai Use code: UGDATASCIDC for 20% off.

    August 1

  • ams244

    Has Dr. Nay worked with the Understanding Risk Group from the World Bank?

    July 30

  • Majid A.

    hi john. vanderbilt alum here from engineering as well. what dept. in the engineering school are you in?

    i'm just wondering how your work fits in with engineering academics -- a certain discipline (not research).

    July 21

    • Doug_S

      I'm wondering how it fits in with David Schum's (George Mason University) work on developing a science of evidence and Judea Pearl's work on causation and evidence. This should be a lively conversation.

      July 21

Our Sponsors

People in this
Meetup are also in:

Sign up

Meetup members, Log in

By clicking "Sign up" or "Sign up using Facebook", you confirm that you accept our Terms of Service & Privacy Policy