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 (http://www.statistics.com/?utm_source=dsdc-meetup&utm_medium=meetup&utm_campaign=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 (http://predictgov.com/), 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 (http://johnjnay.com/predMarket/), software for forecasting drought globally with satellite data (http://johnjnay.com/forecastVeg/), computational models predicting human cooperation (http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0155656), software for automatically estimating models of decision-making (http://www.informs-sim.org/wsc15papers/272.pdf), natural language processing of law and policy (https://dl.dropboxusercontent.com/u/108075111/Nay_2016_gov2vec.pdf), and machine learning for predicting and understanding law-making (http://arxiv.org/abs/1607.02109). 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 (http://johnjnay.com/).

Sponsors:

This event is sponsored by the George Washington Business School MS in Business Analytics Program (http://business.gwu.edu/programs/specialized-masters/m-s-in-business-analytics/academic-program/), Statistics.com (http://bit.ly/12YljkP), Elder Research (http://datamininglab.com/), Novetta (https://www.novetta.com/), PAWGOV (http://www.predictiveanalyticsworld.com/gov/2016/), O'Reilly (http://www.oreilly.com/), Booz Allen Hamilton (https://www.boozallen.com/consulting/strategic-innovation/nextgen-analytics-data-science), and AOL (http://engineering.aol.com/). (Would your organization like to sponsor too? Please get in touch!)